PREFACE

This book describes measurement methods in medicine and biology. While many books on
medical instrumentation cover only hospital instrumentation, this book also encompasses
measurements in the growing fields of molecular biology and biotechnology.
As a first course in bioinstrumentation for undergraduate biomedical engineers, this
book describes measurement methods used in the growing fields of cell engineering, tissue
engineering, and biomaterials. It will also serve as a text for medical and biological personnel
who wish to learn measurement techniques.
Chapter 1 covers the basics of instrumentation systems, errors, and the statistics
required to determine how many subjects are required in a research study. Because many
biomedical engineers are not specialists in electronics, chapter 2 provides the necessary
background in circuits, amplifiers, filters, converters, and signal processing.
Chapter 3 describes clinical measurements of molecules, such as oxygen and glucose,
to biotechnology measurements such as DNA sequencing. For the fields of biomaterials and
tissue engineering, chapter 4 covers measurements on polymers, using surface analysis,
protein adsorption and molecular size.
Measurements on blood components are the commonest measurements on cells, as
described in chapter 5. Cells are counted and identified using changes in impedance and light
scattering. Chapter 6 covers cellular measurements in biomaterials and tissue engineering,
such as cellular orientation, rolling velocity, pore size, deformation, shear stress, adhesion,
migration, uptake, protein secretion, proliferation, differentiation, signaling and regulation.
Chapter 7 describes measurements on the nervous systemaction potentials, EEG,
ERG, EOG, EMG and audiometryand brain imaging using X ray, CT, MRI, nuclear
imaging, SPECT, PET, and biomagnetism. Chapter 8 covers heart and circulation, with
measurements of cardiac biopotentials, pressures, output, sounds, viabilityas well as blood
flow and pressure in the periphery.
Measurements of pulmonary volume, flow, diffusion, and airway resistance are
described in chapter 9, which also includes kidney clearance, bone mineral, and skin water
loss. Chapter 10 covers measurements of body temperature, fat, and movement.
Each chapter has references for further study as well as homework problems to test
comprehension. A web site contains complete laboratory instructions for 12 laboratories,
examination questions, and quiz questions.
Suggested prerequisite courses are calculus, biology, chemistry, and physics. I would
welcome suggestions for improvement of subsequent editions.
John G. Webster
[email protected]
January 2002

1
Measurement Systems
Kevin Hugo

1.1 Studying biomedical engineering

Edna Jones is a 67-year-old retired teacher who is having difficulty with her vision. She
can't remember when it started, but her eyesight has increasingly blurred over the course
of the last year. It has now become so poor that last week she drove her car off the road.
She hasnt been to a doctor for many years, but now shes decided to see her family physician.
The nurse measured and recorded Ednas vital signs. Although her weight, 90 kg
(198 lb.), was measured with an electronic scale, her height, 173 cm (5 ft. 8 in.), was
measured manually. She marveled at the device that took her blood pressure (118/76) and
pulse (63 beats per minute) simultaneously. Another automated instrument took her body
temperature (98.6 F or 37 C) from her ear in a few seconds, instead of having to place a
thermometer under her tongue for a minute.
Edna related her vision problems and accident to the doctor. She is certain that
she did not suffer any injuries. She finds it strange that the doctor asked about how much
water she drinks, but she admitted that she has been rather thirsty lately and that she
drinks and passes a lot of water. The doctor also noted that that Edna has gained a fair
amount of weight.
The doctor began the physical examination by observing Ednas eyes with an
ophthalmoscope. There were cotton-like blotches on the normally orangepink retina.
The physician then felt her hands and feet; they were cold, and she has lost some of the
sensation of touch in her feet. Finally, samples of urine and blood were taken.
It takes several minutes for the laboratory at the clinic to analyze the urine and
blood samples; they use an automated instrument to measure the blood sugar (glucose) in
each. The doctor is concerned about the results, and orders an electrocardiogram and further tests on the blood and urine samples before returning to Ms. Jones to discuss her
condition.

1-1

1-2

Chapter 1 Measurement Systems

1.1.1 Scales of biological organization

The patient in the above story has noninsulin-dependent diabetes mellitus (NIDDM), a
disease in which afflicted persons have high levels of blood glucose. Diabetes mellitus is
a very common chronic disease with many complications, and is one of the most expensive diseases to treat in the United States. However, with early detection and diligent
treatment, the risk of complications can be reduced (Diabetes, 1993).
In diagnosing the patients disease and managing the future treatment, it is necessary to take measurements using biomedical instruments, which is the subject of this
book. One way to classify the various types of bioinstrumentation is to consider the level
of biological organization involved in taking the measurement. For example, it may involve molecules, cells, tissues, organ systems, or the entire body. The chapters of this
book are presented in the order of these biological levels.
This chapter discusses how and why measurements are taken. For example,
every patient has vital signs taken not only because they may be useful in analyzing the
problem that the patient is concerned with, but also because they indicate that persons
general health. We must also understand that the results of measurements will vary due to
errors in taking the measurement as well as individual differences.
An automated thermometer measured Ms. Jones temperature in just several
seconds. This is accomplished by using a temperature-sensing element as well as electronics to derive the temperature based on a mathematical algorithm. This type of electronic signal processing is explained in Chapter 2.
Measurements of concentrations of glucose and other molecules are performed
by spectrophotometry, as described in Chapter 3. Ms. Jones may have coronary artery
disease and need an implanted graft vessel. Chapter 4 describes the electron microscopes
used to test proposed biomaterials and the adhesion of protein molecules to them.
Methods used to analyze the blood sample that was drawn are described in
Chapter 5. Diabetes causes a number of complications. In order to study the urinary casts
(cylindrical protein molds) passed in the urine, Chapter 6 describes the optical microscopes used. Ms. Jones reported a loss of feeling in her legs. Chapter 7 describes the
nervous system and the test used to diagnose nerve disease.
Chapter 8 describes the electrocardiogram, which is used to diagnose Ms. Jones
cardiac problems, as well as methods to obtain blood pressure measurements. Ms. Jones
kidney function is diagnosed by measuring creatinine clearance, as described in Chapter
9. NIDDM patients are frequently overweight. Chapter 10 describes measurements on the
body, including measurements of body fat.
1.1.2 Fields of biomedical engineering
Biomedical engineering is a cross-disciplinary field that incorporates engineering, biology, chemistry, and medicine. This involves both the acquisition of new information and
technology, but also of the development of new devices, processes, and algorithms in
regards to biological systems.
Table 1.1 shows that instrumentation can be classified by the field of biomedical
engineering responsible for its design. Each of these fields is briefly described in the fol-

1.1 Studying Biomedical Engineering

1-3

lowing paragraphs in this section. The Whitaker Foundation describes various fields of
biomedical engineering at http://www.whitaker.org/glance/acareer.html.
Bioinstrumentation
Biomaterials
Biomechanics
Biosignals
Biosystems
Biotransport
Cellular engineering
Clinical engineering
Tissue engineering
Rehabilitation engineering

Table 1.1 Biomedical engineers work in a variety of fields.

Bioinstrumentation applies the fundamentals of measurement science to biomedical instrumentation. It emphasizes common principles and unique problems associated with making measurements in living systems. A physiological variable originates
from a molecular, cellular or systemic process whose nature may be described by a mechanical, electrical, chemical, optical or other event. A variable must be carefully specified before being measured. The sensor should be designed to minimize the disturbance
to the measured variable and its environment, comply with the requirements of the living
system, and maximize the signal-to-noise ratio, i.e. the clarity of the signal. The signal,
typically after being converted into an electrical form, is then conditioned using linear
and nonlinear signal processing, and delivered to an appropriate output device. Bioinstrumentation includes methods for obtaining invasive and noninvasive measurements
from the human body, organs, cells and molecules; electronic instrumentation, principles
of analog and digital signal processing, and typical output devices. It includes measurement concepts such as accuracy, reproducibility, noise suppression, calibration methods,
and safety requirements.
Biomaterials is the application of engineering materials to the production of
medical devices and diagnostic products. In the last decade, advances in biology, especially at the molecular level, have lead to the design and development of new classes of
materials derived from natural sources. These include molecularly engineered materials,
hybrid materials and devices, biomimetic or synthetic biological materials, and other biologically related materials. Biomaterials covers current and traditional applications for
biologically and pharmacologically active materials as well as materials used in such
applications as tissue and organ engineering, diagnostic products, and drug delivery.
Biomechanics covers the behavior of biological tissues and fluids to incorporate
complexities ranging from nonlinear viscoelasticity to non-Newtonian flows. Biomechanics includes both biofluid mechanics and biosolid mechanics at the molecular, cellular,
tissue and organ-system levels. Molecular and cellular biomechanics is integrated with
the macroscopic behavior of tissues and organs. This is put into practice by the modeling
of various biological systems. An example of biomechanics is ergonomics, wherein everyday devices, such as chairs and desks, are designed to reduce stress and injury to the
body.

1-4

Chapter 1 Measurement Systems

Biosignals covers time series analysis in biomedical studies and concentrates on

use of data to uncover the nature of the underlying phenomena, the mechanisms of signal
production, and the fundamental origins of the variability in the signal. Standard methods
of signal analysis have been used for characterizing rather than for elucidating the
mechanisms. Biosignals treats signals from a variety of sources; standard methods of
signal analysis, including transform and statistical techniques and their relationships;
methods of signal generation, fractal analysis methods for signals with different characteristics, methods for analyzing chaotic signals, and approaches to reviewing data to determine or distinguish among possible origins.
Biosystems, which uses modern biology, has given us the tools to identify and
characterize molecules and cells, the fundamental building blocks of organ systems. Understanding how molecules and cells function in tissues, organs and organ systems is the
domain of biological systems analysis. Whereas molecular biology has focused on reductionism, biomedical engineering is in a unique position to move biology to the next frontier where synthesis will lead to an understanding of the function of complex systems.
Biosystems analysis integrates properties of biological systems with current tools of systems analysis.
Biotransport phenomena covers transport processes from the organ to the subcellular level. Quantitative understanding of biological processes is based on the fundamentals of the complementary processes of mass, momentum and energy transport.
Transport of ions, substrates, proteins, viruses and cells is a central issue for the quantitative understanding of biological systems. It builds an integrated picture of the convection,
diffusion, permeation, and reaction kinetics in the circulation, through capillary beds, and
through the tissues of the body. While using the fundamentals of transport phenomena for
understanding membrane transport, cellular and tissue energetics, enzymatic regulation
and metabolism, it also considers how data from microsensors, tissue samples or regional
concentrations over time can be acquired, quantified and interpreted. Transport phenomena considers methods suitable for interpreting data from intact systems.
Cellular engineering develops and communicates quantitative biochemical and
biophysical design principles that govern cell function. An engineering perspective integrates underlying molecular mechanisms. Relevant topics are fundamental kinetics, mechanics, and transport processes that use calculus, differential equations and cell biology.
Models analyze cell functions, such as metabolism, signaling and regulation, biomolecular uptake and secretion, proliferation, adhesion, migration, and differentiation characterizing molecular and cellular properties. Techniques from modern biochemistry and molecular biology alter these parameters in order to test model predictions and demonstrate
how the underlying design principles can be applied to manipulate cell function.
Clinical engineering deals with managing diagnostic and laboratory equipment
in the hospital. Clinical engineers work with health care workers to determine equipment
needs, search for optimal equipment, specify equipment, perform incoming inspection,
train health care workers in proper operation, provide inventory of equipment, decide
whether to perform maintenance in house or purchase vendor contracts, and perform
safety inspections.
Rehabilitation engineering is a newer area of biomedical engineering. A rehabilitation engineer works directly with patients such as disabled individuals to assist in
achieving a better standard of life. The rehabilitation engineer modifies or designs new

1.1 Studying Biomedical Engineering

1-5

equipment for such an individual. A rehabilitation engineer could be involved with an

individual requiring a prosthetic limb, by designing the limb to suit the individuals personal needs. Such a limb would be required of an individual who has lost a leg, but still
desires to run on his or her own power.
1.1.3 Fields of biological science
Table 1.2 shows that instrumentation can be organized by the biological discipline that
utilizes it. Each of these disciplines can be divided into several subspecialties. For example, medicine can be subdivided into anatomy, anesthesiology, biomolecular chemistry,
biostatistics and medical informatics, family medicine, medical microbiology and immunology, medical physics, neurology, neurophysiology, obstetrics-gynecology, oncology,
ophthalmology, pathology and laboratory science, pediatrics, physiology, preventive
medicine, psychiatry, radiology, rehabilitation medicine, and surgery.
Agriculture - Soil monitoring
Botany - Measurements of metabolism
Genetics - Human genome project
Medicine - Anesthesiology
Microbiology - Tissue analysis
Pharmacology - Chemical reaction monitoring
Veterinary science - Neutering of animals
Zoology - Organ modeling

Table 1.2 Biomedical engineers work in a variety of disciplines. One example of instrumentation is
listed for each discipline.

1.1.4 Workplace environments for biomedical engineers

Biomedical Engineers may become a part of the workplace in a variety of different areas,
as shown in Table 1.3. Industry has a demand for biomedical engineers for development
of new technologies and also for testing the theories that others develop. They could also
be technology representatives for existing products. Positions in the government often
deal with testing of proposed products for safety and reliability, to safeguard the public
from harm. In a clinical setting, the biomedical engineer may be the only individual with
working knowledge of some of the instruments, and is needed to allow for their efficient
use. Research institutions are also the place of developing new ideas, and often involve
the dissemination of techniques and practices to others to further the general understanding.
Industry
Government
Clinical institutions
Academic research

Table 1.3 Biomedical engineers may work in a variety of environments.

1-6

Chapter 1 Measurement Systems

1.2 The need for bioinstrumentation

1.2.1 The scientific method
Figure 1.1 shows that researchers often need to use instrumentation to obtain data as part
of the scientific method. For example, we might hypothesize that exercise reduces high
blood pressure yet experimentation and analysis are needed to support or refute the hypothesis. Experiments are normally performed multiple times. Then the results can be
analyzed statistically to determine the probability (hopefully less than 0.05) that the results might have been produced by chance. Results are reported in scientific journals with
enough detail so that others can replicate the experiment to confirm them.

Problem
statement

Review
prior work

State
hypothesis

Perform
experiment

Design
further
experiments

Analyze
data

More
experiments
necessary

Final
conclusion
Problem
solved

Figure 1.1 In the scientific method, a hypothesis is tested by experiment to determine its validity.

1.2.2 Clinical diagnoses

Figure 1.2 shows that physicians often need instrumentation to obtain data as part of the
scientific method. For example, a physician obtaining the history of a patient with a complaint of poor vision would list diabetes as one possibility on a differential diagnosis.

1.2 The Need for Bioinstrumentation

Chief
complaint

Obtain
history

List the
differential
daignosis

Examination
and tests

Treatment
and
evaluation

Select
further tests

Use data to
narrow the
diagnosis

Final
diagnosis

More than
one likely

1-7

Only one
likely

Figure 1.2 The physician obtains the history, examines the patient, performs tests to determine the
diagnosis and prescribes treatment.

1.2.3 Feedback in measurement systems

Figure 1.3 shows that the measurand is measured by a sensor, which converts the variable
to an electrical signal, which can undergo signal processing. Sometimes the measurement
system provides feedback through an effector to the subject. As an example, light flashes
stimulate the eye when measuring visual evoked responses from electrodes placed over
the visual cortex of the brain.
Outputs
Measurand

Sensor

Effector

Signal
conditioning

Feedback

Signal
processing

Data
storage

Data
displays

Data
communication

Figure 1.3 A typical measurement system uses sensors to measure the variable, has signal processing and display, and may provide feedback.

Figure 1.4(a) shows that a patient reading an instrument usually lacks sufficient
knowledge to achieve the correct diagnosis. Figure 1.4(b) shows that by adding the clinician to form an effective feedback system, the correct diagnosis and treatment result.
In certain circumstances, proper training of the patient and a well-designed instrument can lead to self-monitoring and self-control, one of the goals of bioinstrumentation. This is shown in Figure 1.5. An example of such a situation is the day-to-day monitoring of glucose by people suffering from diabetes. Such an individual will contact a
clinician if there is an alert from the monitoring instrument.

1-8

Chapter 1 Measurement Systems

Patient

Patient

Instrument

Instrument

Clinician

(a)

(b)

Figure 1.4 (a) Without the clinician, the patient may be operating in an ineffective closed loop
system. (b) The clinician provides knowledge to provide an effective closed loop system.

Clinician

Abnormal
readings

Patient

Instrument

Figure 1.5 In some situations, a patient may monitor vital signs and notify a clinician if abnormalities occur.

1.3 Instrumentation
1.3.1 Measurands
The measurand is the measured quantity. It may be voltage as when measuring the electrocardiogram, pressure as when measuring the blood pressure, force as when measuring
weight, or other quantities. Table 1.4 lists some common biomedical measurands.
Measurands can be obtained by invasive or noninvasive methods. For example,
if we want to accurately measure the blood pressure in the aorta in vivo (i.e. in a living
organism), we must place a pressure sensor on the tip of a catheter (a long narrow tube),
cut into a peripheral vessel, and advance the catheter to the aorta. An alternative invasive
method is to place the pressure sensor at the distal end of the catheter outside the body.

1.3 Instrumentation

1-9

Measurement

Range

Frequency, Hz

Method

Blood flow

1 to 300 mL/s

0 to 20

Blood pressure
Cardiac output
Electrocardiography
Electroencephalography
Electromyography
Electroretinography
pH
pCO2
pO2
Pneumotachography
Respiratory rate
Temperature

0 to 400 mmHg
4 to 25 L/min
0.5 to 5 mV
5 to 300 V
0.1 to 5 mV
0 to 900 V
3 to 13 pH units
40 to 100 mmHg
30 to 100 mmHg
0 to 600 L/min
2 to 50 breaths/min
32 to 40 C

0 to 50
0 to 20
0.05 to 150
0.5 to 150
0 to 10000
0 to 50
0 to 1
0 to 2
0 to 2
0 to 40
0.1 to 10
0 to 0.1

Electromagnetic or ultrasonic
Cuff or strain gage
Fick, dye dilution
Skin electrodes
Scalp electrodes
Needle electrodes
Contact lens electrodes
pH electrode
pCO2 electrode
pO2 electrode
Pneumotachometer
Impedance
Thermistor

Table 1.4 Common medical measurands.

A noninvasive method to measure blood pressure is to place a pressure cuff

around the arm and listen for the Korotkoff sounds due to turbulence of the blood under
the cuff. Urine can be analyzed noninvasively by collecting a sample and analyzing it in
vitro (maintained in an artificial environment outside the living organism) in the clinical
chemistry laboratory. Weight and electrocardiography are also examples of noninvasive
measurement methods.
1.3.2 Sensors
A variety of sensors have been adapted from industry for use in measuring biomedical
variables such as pressure, flow, and temperature. Table 1.5 shows the specifications for
a typical blood pressure sensor. Note that several specifications are designed to prevent
patient electrocution.
Specification

Value

Pressure range
Overpressure without damage
Maximum unbalance
Linearity and hysteresis
Risk current at 120 V
Defibrillator withstand

30 to +300 mmHg
400 to +4000 mmHg
75 mmHg
2% of reading or 1 mmHg
10 A
360 J into 50

Table 1.5 Sensor specifications for blood pressure sensors are determined by a committee composed of individuals from academia, industry, hospitals, and government.

Linearity is the degree to which the sensor stays within its theoretical output for
some input. Section 1.3.3 discusses linearity in more detail. Hysteresis is a phenomenon
in which increasing the pressure the sensor gives one output but decreasing the pressure

1-10 Chapter 1 Measurement Systems

to the same previous value gives a different output, as shown in Figure 1.6. Risk current
is the amount of current flowing through the sensor at 120 V. Defibrillator withstand is
the amount of energy (in joules) the sensor can withstand if a defibrillator is used while
the sensor is taking measurements.
Sensor
signal

Measurand

Figure 1.6 A hysteresis loop. The output curve obtained when increasing the measurand is different
from the output obtained when decreasing the measurand.

Optimal sensors are designed specifically for the desired measurand, immune to
interference from other variables, and have high gain as shown in Figure 1.7(b). A sensor
with low gain, as shown in Figure 1.7(a), would not change its signal as much for a small
change in the measurand as one with a high gain.
Most sensors are analog and provide a continuous range of amplitude values for
output, as shown in Figure 1.8(a). Other sensors yield the digital output shown in Figure
1.8(b). Digital output has poorer resolution, but does not require conversion before being
input to digital computers and is more immune to interference. Digital and analog signals
will be examined with more detail in Chapter 2.
Sensor
signal

Sensor
signal

Measurand
(a)

Measurand
(b)

Figure 1.7 (a) A low-sensitivity sensor has low gain. (b) A high sensitivity sensor has high gain.
Amplitude

Amplitude

Time
(a)

Time
(b)

Figure 1.8 (a) Analog signals can have any amplitude value. (b) Digital signals have a limited
number of amplitude values.

1.3 Instrumentation

1-11

1.3.3 System characteristics

Specific characteristics
Bioinstrumentation should be designed with a specific signal in mind. Table 1.6 shows a
few specifications for an electrocardiograph (a device that records the electrical activity
generated by the heart.) The values of the specifications, which have been agreed upon by
a committee, are drawn from research, hospitals, industry, and government.
Specification

Value

Input signal dynamic range

Dc offset voltage
Slew rate
Frequency response
Input impedance at 10 Hz
Dc lead current
Return time after lead switch
Overload voltage without damage
Risk current at 120 V

5 mV
300 mV
320 mV/s
0.05 to 150 Hz
2.5 M
0.1 A
1s
5000 V
10 A

Table 1.6 Specification values for an electrocardiograph are agreed upon by a committee.

The electrocardiogram (ECG), the recording of the electrical activity of the heart,
can have a voltage value of anywhere between 0.5 mV and 5 mV (Table 1.4). The electrocardiograph should be able to handle this range, therefore its dynamic input signal
range is designed to handle signals from 5 mV to 5 mV. If the measurand exceeds the
input range of the instrument, the output will saturate and cause an error. Figure 1.9
shows an input signal that exceeds the 5 mV input range (a) and the resulting amplified
signal (b).
Direct current (dc) offset voltage is the amount a signal may be moved from its
baseline and still be amplified properly by the system. Figure 1.10 shows an input signal
without (a) and with (b) offset.
Slew rate is the maximum rate at which the system can observe a changing voltage per unit time without distortion. In the specifications above, if the signal changes
more than 320 mV/s then the resulting output signal will be distorted.
The ECG signal is composed of components that vary in frequency from 0.05
Hz to 150 Hz (Table 1.4) The electrocardiograph is designed to accommodate this as is
shown by its frequency response range. The frequency response of a device is the range
of frequencies of a measurand that it can handle. Frequency response is usually plotted as
gain versus frequency. Figure 1.11 shows the plotted frequency response of this electrocardiograph. Frequency response is discussed in greater detail in Section 2.2.6.
Input impedance is the amount of resistance the signal sees coming into the electrocardiograph. It is desirable to have very high input impedances, in this case 2.5 M,

1-12 Chapter 1 Measurement Systems

so that the majority of the signal will enter the system to be processed and not be lost in
the skin impedance.
Amplitude
5 mV

Time

Dynamic
range

5 mV
(a)
Amplitude
1V

Time
1 V

(b)

Figure 1.9 (a) An input signal that exceeds the dynamic range. (b) The resulting amplified signal is
saturated at 1 V.
Amplitude

Time
(a)
Amplitude

Dc offset
Time
(b)

Figure 1.10 (a) An input signal without dc offset. (b) An input signal with dc offset.

Dc lead current is the amount of current that flows through the leads to the patient being observed by the electrocardiograph. It is important that current flowing to the
patient not exceed very miniscule amounts (i.e. on the order of microamperes). This is
because larger currents can polarize electrodes, resulting in a large offset voltage at the
amplifier input.

1.3 Instrumentation

1-13

When recording the electrocardiogram, several lead configurations can be used.

When switching between the different lead configurations, the electrocardiograph should
take no more than 1 s to adjust itself to correctly display the signal again. Lead configurations, as well as the ECG in general, are discussed in greater detail in Chapter 8.
Overload voltage is the maximum voltage the electrocardiograph can withstand
without damage from defibrillators. A defibrillator is a device that attempts to jump
start the heart by using a large amount of voltage (i.e. on the order of a few thousand
volts.) Defibrillators have two paddles that are placed on the patients chest. When a large
voltage difference exists between the two paddles, current flows through the body from
one paddle to the other. This current is what causes the heart to begin beating again.
Amplitude
1.0

0.1
0.05 Hz

Frequency

150 Hz

Figure 1.11 Frequency response of the electrocardiograph.

When placing leads on a patient that will be hooked up to instrumentation, it is

of paramount importance to make sure the instrumentation does not shock the patient.
Risk current is the amount of current present in the leads due to leakage from the 120 V
of the electric power system.
General Characteristics
Another characteristic that all systems have is whether or not they are linear. Linearity is
highly desirable for simplifying signal processing. For example, if you measure some
input (measurand) you get some output. Now if you double that input you would expect
to double the output. This is not the case for nonlinear systems. Doubling the input in a
nonlinear system might yield triple the output. Figure 1.12 shows the input/output relation of a linear system (a) and of a nonlinear system (b).

1-14 Chapter 1 Measurement Systems

Output

Output

Input
(a)

Input
(b)

Figure 1.12 (a) A linear system fits the equation y = mx + b. Note that all variables are italic. (b) A
nonlinear system does not fit a straight line.

Finally, Figure 1.13 shows that all bioinstrumentation observes the measurand
either continuously or periodically. Continuous signal measurement is most desirable
because no data are lost. However, computer-based systems require periodic measurements since by their nature, computers can only accept discrete numbers at discrete intervals of time.
Amplitude

Amplitude

Time
(a)

Time
(b)

Figure 1.13 (a) Continuous signals have values at every instant of time. (b) Discrete-time signals
are sampled periodically and do not provide values between these sampling times.

1.3.4 Outputs
Bioinstrumentation can direct its output three ways: display(s), storage, and communication. A display is simply a monitor or some other device attached to the instrumentation
that gives the user a sense of what the input is doing. The term display is a bit misleading
since output does not necessarily have to be visible (e.g. a stethoscope uses sound for its
output.) Storage is increasingly common as computers are integrated into bioinstrumentation. Storage usually takes the form of saving data onto a hard drive or some other type of
permanent record. Communication involves transmitting the output to some other location for analysis. Modern imaging systems make extensive use of communication via
computer networks and wireless transmission.
1.3.5 Panels and series
Certain groups of measurements are often ordered together because they are very commonly used or because they are related. This may occur even if the measurements are

1.3 Instrumentation

1-15

based on different principles or are taken with different sensors. Table 1.7 is an example
of one of these groups of measurements, which are called panels or series.
Laboratory test

Typical value

Hemoglobin
Hematocrit
Erythrocyte count
Leukocyte count
Differential count

13.5 to 18 g/dL
40 to 54%
4.6 to 6.2 106/L
4500 to 11000/L
Neutrophil 35 to 71%
Band 0 to 6%
Lymphocyte 24 to 44%
Monocyte 1 to 10%
Eosinophil 0 to 4%
Basophil 0 to 2%

Table 1.7 Complete blood count for a male subject.

Hemoglobin is the protein which caries oxygen in the bloodstream. Hematocrit

is the percent of solid material in a given volume of blood after it has been centrifuged.
An erythrocyte is a red blood cell. A leukocyte is a white blood cell. The differential
count tells how many of each type of white blood cell there are in one microliter of blood.
Unusual values for different leukocytes can be indicative of the immune system fighting
off foreign bodies.

1.4 Errors in measurements

1.4.1 Sources of error
When we measure a variable, for example the voltage of the electrocardiogram, we seek
to determine the true value, as shown in Figure 1.14(a). This true value may be corrupted
by a variety of errors. For example movement of electrodes on the skin may cause an
undesired added voltage called an artifact. Electric and magnetic fields from the power
lines may couple into the wires and cause an undesired added voltage called interference,
as shown in Figure 1.14(b). Thermal voltages in the amplifier semiconductor junctions
may cause an undesired added random voltage called noise. Temperature changes in the
amplifier electronic components may cause undesired slow changes in voltage called drift.
We must evaluate each of these error sources to determine their size and what we can do
to minimize them. Usually we try to minimize the largest error, since that is the weakest
link in the measurement.

1-16 Chapter 1 Measurement Systems

(b)

(a)

Figure 1.14 (a) Signals without noise are uncorrupted. (b) Interference superimposed on signals
causes error. Frequency filters that can be used to reduce noise and interference are discussed in
Chapter 2.

(a)

(b)

(c)

Figure 1.15 (a) Original waveform. (b) An interfering input may shift the baseline. (c) A modifying
input may change the gain.

Figure 1.15(b) shows that an interfering input may be added to the original
waveform and shift it away from the baseline. Figure 1.15(c) shows that a modifying
input may multiply the original waveform to change its gain.
1.4.2 Accuracy and precision
Resolution is the smallest incremental quantity that can be reliably measured. For example, a voltmeter with a larger number of digits has a higher resolution than one with fewer
digits. A speedometer with more tick marks between each numbered speed has a higher
resolution than one with no tick marks. However, high resolution does not imply high
accuracy.
Figure 1.16 shows that precision is the quality of obtaining the same output from
repeated measurements from the same input under the same conditions. High resolution
implies high precision.
Repeatability is the quality of obtaining the same output from repeated measurements from the same input over a period of time.
Accuracy is the difference between the true value and the measured value divided by the true value. Figure 1.17(a) shows low accuracy and high precision. Figure
1.17(b) shows high accuracy and high precision.
Obtaining the highest possible precision, repeatability, and accuracy is a major
goal in bioinstrumentation design.

1.4 Errors in Measurements

(a)

(b)

Figure 1.16 Data points with (a) low precision and (b) high precision.

(a)

(b)

Figure 1.17 Data points with (a) low accuracy and (b) high accuracy.

1.4.3 Calibration

1-17

1-18 Chapter 1 Measurement Systems

Measuring instruments should be calibrated against a standard that has an accuracy 3 to
10 times better than the desired calibration accuracy. The accuracy of the standard should
be traceable to the National Institute of Standards and Technology http://www.nist.gov/.
If the instrument is linear, its output can be set to zero for zero input. Then a
one-point calibration defines the calibration curve that plots output versus input. Figure
1.18 shows that if the linearity is unknown, a two-point calibration should be performed
and these two points plus the zero point plotted to ensure linearity. If the resulting curve
is nonlinear, many points should be measured and plotted to obtain the calibration curve.
If the output cannot be set to zero for zero input, measurements should be performed at zero and full scale for linear instruments and at more points for nonlinear instruments.
Calibration curves should be obtained at several expected temperatures to determine temperature drift of the zero point and the gain.
Output

Output

Input
(a)

Input
(b)

Figure 1.18 (a) The one-point calibration may miss nonlinearity. (b) The two-point calibration may
also miss nonlinearity.

1.5 Statistics
1.5.1 Distributions
Mean
If we make n measurements of x, for example of the weights of a population, we may
wish to report an estimate of them in a condensed way. The simplest statistic is the estimated sample mean

x=

where i = 1, 2,n.

xi
n

(1.1)

1.5 Statistics

1-19

Standard Deviation
A measure of the spread of data about the mean is the estimated sample standard deviation
2
( xi x )

s=

n 1

(1.2)

The standard deviation of the mean s is

x=

s
n 1

(1.3)

Example 1.1 If Person 1 weighs 68 kg, Person 2 weighs 90 kg and Person 3 weighs 95
kg, find the mean, the standard deviation, and the standard deviation of the mean.
Here n = 3. The mean of the data is

x=

xi
n

68 + 90 + 95
= 84.3 kg
3

The standard deviation of the data is

s=

2
(x i x )

n 1

(68 84.3)2 + (90 84.3)2 + (95 84.3)2

3 1

= 14.4 kg

The standard deviation of the mean is

x=

s
n 1

14.4
3 1

= 10.2 kg

Gaussian Distribution

The spread (distribution) of data may be rectangular, skewed, Gaussian, or other. The
Gaussian distribution is given by

f (X ) =

2
2
e ( X ) (2 )

(1.4)

1-20 Chapter 1 Measurement Systems

where is the true mean and is the true standard deviation of a very large number of
measurements. Figure 1.19 shows the familiar bell curve shape of the Gaussian distribution.

Frequency

xs

Estimated mean

standard deviation sx

Population standard
deviation

Mean

Figure 1.19 For the normal distribution, 68% of the data lie within 1 standard deviation. By
measuring samples and averaging, we obtain the estimated mean x s , which has a smaller standard
deviation sx. is the tail probability that xs does not differ from by more than .

Poisson Probability

The Poisson probability density function is another type of distribution. It can describe,
among other things, the probability of radioactive decay events, cells flowing through a
counter, or the incidence of light photons. The probability that a particular number of
events K will occur in a measurement (or during a time) having an average number of
events m is
p ( K ; m) = e

mK

K!

The standard deviation of the Poisson distribution is

Poisson distribution.

(1.5)

m . Figure 1.20 shows a typical

1.5 Statistics
p

1-21

0.2

0.1

0.0
0

2 K

Figure 1.20 A typical Poisson distribution for m = 3.

Example 1.2 Assume that the average number of cells passing through a counter in one
sampling period is 10. Calculate the probability that 15 cells will be counted in a sampling period.
Since 10 is the average number counted cells, m = 10. We want to find the probability of 15 cells being counted so K = 15.

p ( K ; m) = e

mK = e 10 1015 = 0.035

K!

15!

1.5.2 Statistical estimations

Suppose we want to know the average weight of adults. It is not feasible to weigh every
single adult and then take the average of all the weights. All adults are called the population. Instead, we decide to take a small fraction of the adults, say 1 out of every 1000, and
average these weights. This small fraction is called our sample population. Now we have
an average weight for our sample population. We want to know if our sample population
average weight is a good estimation of the population average weight. There are two
ways for us to find out.
If we represent our sample population average weight as a single number, xs ,
then we have made a point estimation of the population average weight. Lets call the
population average weight and the difference between and x s is . The standard deviation of the population is . The tail probability that x s will not differ from by more
than is . We can control the probability of how much x s differs from by adjusting
the sample size, N, using the formula
2

N = Z 2

(1.6)
where Z is a multiple of the standard deviation. This number is obtained from a statistical table (Distribution Tables, 2002).

1-22 Chapter 1 Measurement Systems

The confidence is how sure one wants to be that the sample size, N, calculated will actually meet the specifications for

1 percent confidence
2

(1.7)

If we know the standard deviation, , of the population average weight, we can use it.
Often, the standard deviation of a population can be found in the literature or from some
other source. If it cannot, then the sample standard deviation, s, must be used instead.
Also, the Student t distribution, t, must be used instead of Z.
s
N = t2

(1.8)

Example 1.3 Suppose we know that the standard deviation of the adult average weight is
3 kg. Using Eq. (1.6), we will set to 2, which means our estimate does not deviate from
the population average weight by more than 2 kg. Lastly, we want a 95% confidence
that the sample size N we calculate will meet our specifications for .

We calculate = (1 0.95)/2 = 2.5% = 0.025. We look up in a statistical table

(Distribution Tables, 2002) the value for Z.025 and find that it equals 1.96. We can then
find N by using
2

3
N = (1.96) 2 = 8.6 9
2

which indicates that we would need to take the average of nine weights to get an estimation of the population average weight within 2 kg with a confidence of 95%. We could
then say that the average weight of the population was some value, w.
Another way to represent estimations is with a confidence interval. When dealing with confidence intervals, is twice as large because it represents the difference between the upper limit and lower limit instead of between and x s . The sample size
needed for an interval estimation is twice that needed for a point estimation and can be
found with the formula

N = 2Z 2

(1.9)

Using the conditions stated above, we could solve for N and find it equal to approximately 18. We would need 18 samples to be 95% confident that our weight represented
the average weight of the population. We could then report our average weight, w, as

1.5 Statistics

1-23

[Lower limit on average weight] w [Upper limit on average weight]

In the event that we do not know the standard deviation of the population, we can use the
sample standard deviation, s, and the Student t distribution t.
1.5.3 Hypothesis testing

In hypothesis testing, there are two hypotheses. H0, the null hypothesis, is a hypothesis
that assumes that the variable in the experiment will have no effect on the result and Ha is
the alternative hypothesis that states that the variable will affect the results. For any population, one of the two hypotheses must be true. The goal of hypothesis testing is to find
out which hypothesis is true by sampling the population.
In reality, H0 is either true or false and we draw a conclusion from our tests of
either true or false. This leads to four possibilities, as shown in Table 1.8.
Usually, the probability of making a Type I error is designated as and the
probability of making a Type II error is designated as . Common values for , the significance level, are 0.01, 0.05, and 0.1. The statistical power of a test is the probability
that H0 will be rejected when it is false. Statistical power is given by 1 . Common
power levels are above 0.8, indicating that when H0 is false it will be correctly rejected
more than 80% of the time.
Conclusion
Accept H0
Reject H0

Real situation
H0 true
Correct decision
Type I error, p =

Ha true
Type II error, p =
Correct decision

Table 1.8 The four outcomes of hypothesis testing.

Many scientific studies involving a hypothesis test will report the p value of the
test. A very small p value means that the null hypothesis is unlikely to be true. When p is
below an arbitrary cut-off value, e.g. 0.05, the result is called statistically significant. A
study that reports p = 0.05 means that since the result obtained would happen only once
in 20 times if H0 were true, then H0 should be rejected.
Test result
Negative
Positive

Has condition?
No
True negative (TN)
False positive (FP)

Yes
False negative (FN)
True positive (TP)

Table 1.9 Equivalent table of Table 1.8 for results relating to a condition or disease.

Test results about a condition or disease use the terminology in Table 1.9. In this
case H0 states that an individual does not have a condition or a disease, whereas Ha states
that an individual does have a condition or disease.

1-24 Chapter 1 Measurement Systems

The terms in Table 1.9 are useful for defining measures that describe the proportion of, for example, a disease in a population and the success of a test in identifying it.
Incidence is the number of cases of a disease during a stated period, such as x
cases per 1000 per year.
Prevalence is the number of cases of a disease at a given time such as y cases per
1000. It is all diseased persons divided by all persons.
Prevalence =

TP + FN
TN + TP + FN + FP

(1.10)

Sensitivity is the probability of a positive test result when the disease is present.
Among all diseased persons, it is the percent who test positive.
Sensitivity =

TP
100%
TP + FN

(1.11)

Specificity is the probability of a negative diagnostic result in the absence of the

disease. Among all normal persons, it is the percent who test negative.
Specificity =

TN
100%
TN + FP

(1.12)

Considering only those who test positive, positive predictive value (PPV) is the
ratio of patients who have the disease to all who test positive.
PPV =

TP
100%
TP + FP

(1.13)

Considering only those who test negative, negative predictive value (NPV) is the
ratio of nondiseased patients to all who test negative.
NPV =

TN
100%
TN + FN

(1.14)

In order to estimate if a proposed test is useful or not, we must run trials on a

sample of subjects who are known normal and diseased. We can calculate the number of
subjects required in our trial by consulting books on biostatistics (Dawson-Saunders and
Trapp 1990).
Clinical trials are prospective double blind randomized treatments with additional requirements on sample selection and ethical issues.

1.5 Statistics

1-25

Tests to diagnose disease are not perfect. Figure 1.21 shows that positive test
results may occur for only a portion of the normal population and only a portion of the
diseased population. We set the threshold for the test to maximize true results and minimize false results.
Normal
population

Threshold

True
negative
False positive,
p=

Diseased
population

True
positive

False negative,
p=

Figure 1.21 The test result threshold is set to minimize false positives and false negatives.

The sample size needed to test a hypothesis depends on the level of significance
and power desired. We can increase the threshold to make it very difficult to reject H0 in
order to reduce Type I errors (false positives). But then we risk not rejecting H0 when it
is false and making a Type II error (false negative). If we decrease the threshold, the converse is also true.
For a test comparing the means m1, m2 of two Gaussian populations with known
and equal standard deviations, , the size of each sample needed to test H0: |m1 m2| < ,
Ha: |m1 m2| > , is (Selwyn, 1996, Appendix B)

N = 2 Z + Z

)2

(1.15)

where Z and Z are the values for the standard Gaussian distribution corresponding to,
respectively, the tail probabilities of size and . For example, if = 0.05, = 0.1, and
1 = 0.9, Z = 1.96 and Z = 1.65 and N = 26.06(/)2. If the result of the test suggests

1-26 Chapter 1 Measurement Systems

that we accept H0, then there is a 5% probability that H0 is false. If the result of the test
suggests that we reject H0, then there is a 10% probability that H0 is true.
For a test about the mean m of a normal population, with H0: |m| < and Ha: |m|
> , the sample size needed is

N = Z + Z

)2

(1.16)

where Z and Z have the same meaning as in Eq. (1.15). Kanji (1993) provides a fine
summary of statistical tests.
A final word of caution about N in the above equations. All the sample sizes
calculated are estimates based on several assumptions, some of which may be difficult to
prove. N is not an exact number. Hence, it may be convenient to modify some of the relevant factors involved in the equations in order to seek their effect on N. In any case, a
high value for N indicates the need for pilot studies aimed to reduce the experimental
variability.

1.6 Lifelong learning

Most biomedical engineers have had formal education from a college through a program
of instruction designed by others. This education should continue throughout their career
through a program of instruction designed by themselves. New technology appears constantly and must be learned to promote a productive career. A good place to start is by
reading books, which review a large amount of material, present it in an organized fashion, and provide references for further study. To review books and conference proceedings, go to a technical library electronic card catalog or www.amazon.com and search on
biomedical engineering.
To obtain recent information, go to www.bmenet.org/BMEnet/ or
www.mecca.org/BME/BMES/society/, join societies such as the Biomedical Engineering
Society or the Institute of Biomedical Engineering Technology, subscribe to biomedical
engineering journals such as the Annals of Biomedical Engineering, and go to conferences listed there. You can search medical databases such as MEDLINE at
www.ncbi.nlm.nih.gov/PubMed/. You can search patents at http://www.delphion.com/
Commercial databases available at universities include ISI, OVID, and INSPEC. Perform
general searches using search engines such as http://www.google.com/.

1.7 References
Biomedical Engineering: IEEE EMBS [Online] www.eng.unsw.edu.au/embs/index.html.
Dawson-Saunders, B. and Trapp, R. G. 1990. Basic and Clinical Biostatistics. Norwalk,
CT: Appleton & Lange.

1.7 References

1-27

Doebelin, E. O. 1990. Measurement Systems: Application and Design. 4th ed. New York:
McGraw-Hill.
Kanji, G. K. 1993. 100 Statistical Tests. Newbury Park, CA: SAGE Publications.
Nathan, D. M. 1993. Long-term complications of diabetes mellitus. N. Engl. J. Med., 328
(23): 167685.
Distribution Tables 2002. [Online] www.statsoft.com/textbook/stathome.html
The Diabetes Control and Complications Trial Research Group. 1993. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N. Engl. J. Med., 329 (14): 97786.
Selwyn, M. R. 1996. Principles of Experimental Design for the Life Sciences. Boca Raton:
CRC Press.
Webster, J. G. (ed.) 1998. Medical Instrumentation: Application and Design. 3rd ed. New
York: John Wiley & Sons.
The Whitaker Foundation [Online] www.whitaker.org

1.8 Problems
1.1

Compare the scientific method with that of making a diagnosis. List three things
these processes have in common and three differences.
1.2 Give an example of a problem requiring different areas of biomedical engineering
to be integrated in order to solve it.
1.3 Describe why a differential diagnosis is important.
1.4 Apply the steps of diagnosing diseases to diabetes.
1.5 For a system with a continuous, constant drift, explain what type of calibration
would work best.
1.6 Your samples from a population are 1, 1, 3, 5, 5. Estimate the mean x and standard deviation s.
1.7 Your samples from a population are 1, 1, 3, 5, 5. Estimate the standard deviation of
the mean sx.
1.8 For the Poisson probability density distribution, calculate and plot p(K;m) for K = 0,
1, 2, 3, 4, and m = 2.
1.9 An X-ray image has 1000 1000 picture elements (pixels). Calculate the number
of X-ray photons required so that the signal-to-noise ratio (SNR) = (average number of photons)/(standard deviation of the number of photons) = 8.
1.10 The results of test X are based on a threshold; positive results are greater than the
threshold and negative results are less than the threshold. If the threshold is increased, explain how the sensitivity and specificity of test X change.
1.11 Find the prevalence, sensitivity, specificity, positive predictive value, and negative
predictive value for a test where TP = 80, FN = 20, TN = 30, FP =10.
1.12 A test had 90% sensitivity and 80% specificity. If there are 5 people actually affected by the condition and 60 not affected, then find the number of true positives,
true negatives, false positives, and false negatives.

1-28 Chapter 1 Measurement Systems

1.13 A test has 83.3% sensitivity and 91.7% specificity. If the tests report that 15 people
are positive and 57 people are negative, calculate the number of TP, TN, FP, and
FN.
1.14 Express positive predictive value and negative predictive value in terms of the
prevalence, sensitivity, and specificity.
1.15 For the blood pressure of adults, the mean = 120 mmHg and the standard deviation = 5 mmHg. Calculate the number N of adults we must test to obtain a probability p = 0.95 that our sample population average blood pressure x s does not differ from by more than = 3 mmHg.
1.16 Find a magazine or journal article about a new technology in biomedical engineering. Provide the complete reference. Write a one-paragraph summary of what it
may be used for and how it works.
1.17 Find web information about a new technology in biomedical engineering. Provide
the complete reference. Write a one-paragraph summary of what it may be used for
and how it works.

2
Basic Concepts of Electronics
Hong Cao

Medical instruments are widely used in clinical diagnosis, monitoring, therapy, and
medical research. They provide a quick and precise means by which physicians can
augment their five senses in diagnosing disease. These instruments contain electric
components such as sensors, circuits, and integrated circuit (IC) chips. Modern
electronics technology, which includes transistors, ICs, and computers, has
revolutionized the design of medical instruments.
Biomedical engineers should have a fundamental understanding of their
operations and a basic knowledge of their component electric and electronic s ystems.
Using this knowledge provides a better understand the principles of various
measurements or even develop new measurements and instruments.
Electrical engineering is too large a topic to cover completely in one chapter.
Thus, this chapter presents some very basic concepts in several fields of electrical
engineering. It discusses analog components such as resistors, capacitors, and inductors.
It then goes on to basic circuit analysis, amplifiers, and filters. From this it moves to the
digital domain, which includes converters, sampling theory, and digital signal processing.
It then discusses the basic principles of microcomputers, programming languages,
algorithms, database systems, display components, and recorders.

2.1 Electronic components and circuit analysis

2.1.1 Current
An atom contains a nucleus surrounded by electrons. Most of the electrons are tightly
bound to the atom, while some electrons in the outer orbits are loosely bound and can
move from one atom to another. In conductors, there are many free electrons. This
process of electron transfer occurs in random directions in materials.
1

Chapter No. Title

Suppose there is an imaginary plane in a conductor (Figure 2.1(a)). The loosely

bound outer orbital electrons continuously cross this plane. Due to the random direction
of the electron movement, the number of electrons that cross the plane from left to right
equals the number that cross from right to left. Thus, the net flow of electrons is zero.
I

Electron

(a)

(b)

Figure 2.1 Electric current within a conductor. (a) Random movement of electrons generates no
current. (b) A net flow of electrons generated by an external force.

When an electric field is applied across a conductor, it causes a net flow of

electrons in the conductor because the electrons are attracted to the positive side of the
electric field (Figure 2.1(b)). The rate of flow of electrons through a region is called
electric current.
If Q is the amount of charge that passes perpendicularly through a surface with
area A, in time interval t, the average current, Iav, is equal to the charge that passes
through A per unit time.

I av = Q t

(2.1)

The rate at which current flows varies with time as does charge. We therefore can define
the instantaneous current, I, as the differential limit of Eq. (2.1).

I = dQ dt

(2.2)

The unit of current is the ampere (A), which represents a net flow of one coulomb (1 C),
of charge or 6.242 1018 electrons across the plane per second (s).
The electron has a negative charge. When a negative charge moves in one
direction, it yields the same result as a positive charge moving in the opposite direction.
Conventionally, we define the direction of positive charge movement to be the direction
of the electric current. Figure 2.1(b) shows that the direction of current is opposite to that
of the flow of electrons.

Section No. Title

Current can be generated in a circuit by a current source or by a voltage source

and resistor in series.
2.1.2 Voltage and potential

In moving a charge (+ or ) from point A to point B in an electric field, the potential

energy of the charge changes. That change in energy is the work, W, done by the electric
field on the charge. The amount of work is measured in Joules, J. If we let the electric
potential, V, be equal to the potential energy, U, per unit charge, q0, then we can define a
potential difference, or voltage, as

V = VB VA =

WAB
q0

(2.3)

where VB and VA are the potential at points B and A, respectively. The unit of potential is
the volt (V), where 1 J/C = 1 V. If we choose the potential at infinity to be zero, the
absolute potential of a point in an electric field can be defined as the total work per unit
charge that has been done to move the charge from infinity to the point.
It is important that potential difference not be confused with difference in
potential energy. The potential difference is proportional to the change in potential
energy, where the two are related by U = q0V.
From Eq. (2.3), we can determine the work needed to move a charge from A to
B if we know the potential difference between the two points. We are more interested in
potential difference than the absolute potential for a single point. Notice that the potential
is a property of the electric field, whether there is a charge in the field or not.
Voltage can be generated in a circuit by a voltage source or by a current source
and resistor in parallel. Even though they exist, current sources are much less common
than voltage sources in real circuits.
Example 2.1 How much work is needed to move an electron (a charge of 1.61019 C)
from 0 V to 4 V?
Rearranging Eq. (2.3)

WAB

= q0 (VB VA ) = 1.6 10 19 C(4 V 0 V ) = 6.4 10 19 J

2.1.3 Resistors and Ohm's law

When free electrons move in a conductor, they tend to bump into atoms. We call this
property resistance and use a resistor as the electric component implementation. The unit
of resistance is the Ohm (), in honor of Georg Simon Ohm, who discovered what is
now known as Ohms law. Ohms law states that for many materials, when a potential
difference is maintained across the conductor, the ratio of the current density J (current

Chapter No. Title

per unit area) to the electric field that is producing the current, E, is a constant, , that is
independent of the electric field. This constant is called the conductivity of the conductor.
J = E

(2.4)

It is useful to consider Ohms law in the case of a straight wire with cross-sectional area a
and length l, as in Figure 2.2. V = Vb Va is the potential difference maintained across the
wire, which creates an electric field and current in the wire. The potential difference can
be related to the electric field by the relationship
V = El

(2.5)

l
I
Vb

Va

Figure 2.2 A model of a straight wire of length l and cross-sectional area a. A potential difference
of Vb Va is maintained across the conductor, setting up an electric field E. This electric field
produces a current that is proportional to the potential difference.

Since the magnitude of the current density is related as J = E, and current

density J is defined as I/a, the potential difference can be rewritten as
V = (l )J = (l a )I

(2.6)

The quantity (l/a) is defined as the resistance, R, of the conductor. The relationship can
also be defined as a ratio of the potential difference to the current in the conductor
R = (l a ) = V I

(2.7)

Experimental results show that the resistance, R, of a conductor can be rewritten

as
R=

l
a

(2.8)

where , the reciprocal of conductivity , is the resistivity (m), l is the length of the
conductor in meters (m), and a is the cross sectional area of the conductor (m2).
Ohms law is written as
V = IR
(2.9)
where V is the voltage across the resistor and I is the current through the resistor.

Section No. Title

Resistance depends on geometry as well as resistivity. Good electrical

conductors have very low resistivity, where good insulators have very high resistivity.
Resistivity is important in biomedical applications regarding not only electrical
component resistances, but also resistances caused by body substances such as red blood
cells, that may act as an insulator in counting procedures. Common carbon composition
resistors used in electronic circuits have values between 10 to 22 M.
Example 2.2 What is the resistance of a 12.0 cm long piece of copper wire, if the wire
has a resistivity of 1.7108 m and a radius of 0.321 mm?
The cross-sectional area of the wire is

a = r 2 = 0.321 10 3 m

)2 = 3.24 107 m2

The resistance can be found using Eq. (2.8)

R=

0.12 m
l

= 1.7 10 8 m
= 6.30 10 3
7 m 2
a
3
.
24
10

Most electric circuits make use of resistors to control the currents within the
circuit. Resistors often have their resistance value in ohms color coded. This code serves
as a quick guide to constructing electric circuits with the proper resistance. The first two
colors give the first two digits in the resistance value. The third value, called the
multiplier, represents the power of ten that is multiplied by the first two digits. The last
color is the tolerance value of the resistor, which is usually 5%, 10% or 20% (Figure 2.3).
In equation form, the value of a resistors resistance can be calculated using the following
AB 10 C D%

(2.10)

where A is the first color representing the tens digit, B is the second color representing
the ones digit, C is the third color or multiplier and D is the fourth color or tolerance.
Table 2.1 gives the color code for each of these four categories, and their corresponding
digit, multiplier, or tolerance value.

Chapter No. Title

First digit (A)
Multiplier (C)

Second digit (B)

Tolerance (D)

Figure 2.3 The colored bands that are found on a resistor can be used to determine its resistance.
The first and second bands of the resistor give the first two digits of the resistance, and the third
band is the multiplier, which represents the power of ten of the resistance value. The final band
indicates what tolerance value (in %) the resistor possesses. The resistance value written in
equation form is AB10C D%.
Color
Black
Brown
Red
Orange
Yellow
Green
Blue
Violet
Gray
White
Gold
Silver
Colorless

NUMBER
0
1
2
3
4
5
6
7
8
9
1
2

Tolerance (%)

5%
10%
20%

Table 2.1 The color code for resistors. Each color can indicate a first or second digit, a multiplier,
or, in a few cases, a tolerance value.

Example 2.3 What is the resistance value of a resistor where the first band is orange,
second band is blue, third band is yellow, and fourth band is gold?
Using Table 2.1, the first band, orange, represents the first digit, 3. The second
band gives the second digit, 6. The third band, is yellow, 4. The fourth band, or tolerance,
is gold, 5%. Therefore, the resistance value is AB10C D% or 36104 = 360 k, with a
tolerance value of 5% or 18 k.

We define power, the rate at which work is done, as

P = VI

(2.11)

The unit of power is the watt (W). V is the voltage between the two terminals of the
resistor and I is the current through the resistor.
Using Eq. (2.7) we can rewrite Eq. (2.11) as

Section No. Title

P=

V2
= I 2R
R

(2.12)

Although power is most often used to describe the rate at which work can be done by
mechanical objects, in electronics it represents the amount of energy dissipated by a
component. The power lost as heat in a conductor, for example, is called joule heat and is
referred to as a loss of I2R.
2.4 Assume the chest is a cylinder 10 cm in diameter and 40 cm long with a resistivity of
0.8 m. For a voltage of 2 kV during defibrillation, calculate the current and power
dissipated by the chest.
Calculate the current and power by using Eqs. (2.8), (2.9) and (2.11).

l
0.4
= 0.8
= 40.8

a
(0.1) 2
4
V 2000
= 49 A
I= =
R 40.8
P = VI = 2000 49 = 98 kW
R=

2.1.4 Basic circuit analysis

An electric system generally contains many components such as voltage and current
sources, resistors, capacitors, inductors, transistors, and others. Some IC chips, such as
the Intel Pentium, contain more than one million components.
Electric systems composed of components are called networks, or circuits. We
can analyze the performance of simple or complex circuits using one theorem and two
laws of linear circuit analysis.
Superposition theorem: The current in an element of a linear network is the sum
of the currents produced by each energy source.
In other words, if there is more than one source in the network, we can perform
circuit analysis with one source at a time and then sum the results. The superposition
theorem is useful because it simplifies the analysis of complex circuits.
One concept we need to introduce is the polarity of a voltage drop.
Conventionally, if the direction of current flow in a circuit is known or assumed, then the
direction of current flow across a circuit element is from + to . In other words, by
current flowing through an element, a voltage drop is created (as shown by Ohms law)
by that element with the polarity of + to in the direction of current flow (Figure 2.4(a))

Chapter No. Title

R1

R2

+ 10

+ 20

V1

V2

(a)

VS = 30 V
(b)

Figure 2.4 (a) The voltage drop created by an element has the polarity of + to in the direction of
current flow. (b) Kirchhoffs voltage law.

In a circuit, a loop is defined as a closed path containing circuit elements.

Kirchhoffs voltage law (KVL) states

V = 0

(2.13)

or the sum of voltage drops around a loop is zero. In other words, starting at any point on
the closed loop, we can use the voltage law to form an equation by adding each voltage
drop across every element (resistors and sources). We can assume the direction of current,
but if our calculations give a negative result then the actual current flows in the opposite
direction. This law follows from conservation of energy. A charge that moves in a closed
loop must gain as much energy as it loses if a potential can be found at each point in the
circuit.
Figure 2.4(b) shows a circuit to which we can apply Kirchhoffs voltage law. If
we sum the voltages counterclockwise around the loop we get
VS + V1 + V2 = 0
VS = V1 + V2
Notice that the voltage source, VS, has a polarity opposite that of the other voltage drops.
A resistor that is traversed in the direction of the current gives a drop in potential
across the resistor equal to IR, while a resistor that is traversed in the direction opposite
of a current gives an increase in potential across the resistor equal to +IR. This is directly
related to Eq. (2.9), where the direction of current flow across a resistor affects the
potential change across that resistor. Also a voltage source that is traversed from the +
to terminal gives a drop in potential of V, while a voltage source that is traversed from
the to + terminal gives a increase in potential of +V.
In circuit analysis, a node is a junction of two or more branches. Kirchhoffs
current law (KCL) states that at any node

Section No. Title

I = 0

(2.14)

or the sum of currents entering or leaving any node is zero, this follows the law of
conservation of charge. In other words, the currents entering a node must equal the
currents leaving the node (Figure 2.5)

6A

3A

I1

I2

I3

I=?

(b)

(a)

Figure 2.5 (a) Kirchhoffs current law states that the sum of the currents entering a node is 0. (b)
Two currents entering and one negative entering, or leaving.

When we use Kirchhoffs current law, we can arbitrarily label the current
entering the node + and leaving the node . The sum of the currents entering the node in
Figure 2.5(b) is

+3 + 6 I = 0
I =9A
Kirchhoffs voltage law and current law are basic laws for circuit analysis.
There are two types of circuit analysis, each based on each of Kirchhoffs laws.
We assume unknown currents in the loops and set up the equations using
Kirchhoffs voltage law and then solve these equations simultaneously.
We can use Figure 2.4(b) to illustrate loop analysis. We assume a current flows
through the circuit in the direction already indicated in the figure. If we assumed the
current in the other direction, our result would just be negative. Recalling Ohms law, we
have for the sum of the voltages through the clockwise loop
VS = R1I + R2 I
30 = 10 I + 20 I
I = 1A
Now that we know the current through the loop, we can find the voltage drop through
either resistor using Ohms law. In this example, the voltage drop for R1 is 10 V.

10

Chapter No. Title

a

4W

10 V
+

2W
6W

I1

14 V
-

I3

I2

Figure 2.6 Kirchhoffs current law for Example 2.5.

Example 2.5 Find the currents I1, I2 and I3 in the circuit shown in Figure 2.6.
Applying Kirchhoffs current law to the current junction at node c, we have

I1 + I 2 = I3

(1)

All together there are three loops, abcda, befcb, and aefda (the outer loop) to which
Kirchhoffs voltage law can be applied. Applying these Kirchhoffs voltage law to the
two inside loops in the clockwise direction, we have
Loop abcda: 10 V (6 )I1 (2 )I3 = 0

Loop befcb: 14 V 10 V + (6 )I1 (4 )I 2 = 0

(2)
(3)

Expressions (1), (2) and (3) represent three equations with three unknowns, and we can
solve this system of three equations, first by substituting (1) into (2), giving
10 = 8I1 + 2 I 2

(4)

Dividing each term in (3) by 2, we get

12 = 3I1 + 2 I 2
Finally, subtracting (5) from (4), I2 is eliminated, giving
22 = 11I1
I1 = 2 A

(5)

Section No. Title

11

Substituting this value into (5), we find I2 = 3 A., and substituting both these values in
(1), we find I3 = 1 A.
The other type of analysis is nodal analysis. We assume unknown voltages at
each node and write the equations using Kirchhoffs current law. We solve these
equations to yield the voltage at each node. Figure 2.7 shows an example using nodal
analysis.
A

+ 20 W -

+ 20 W -

V
+

30 V

+
20 V

20 W
-

Figure 2.7 Example of nodal analysis.

In Figure 2.7, we are given the voltage at A as 30 V and at B as 20 V. We

assume the voltage at C is V. We apply Kirchhoffs current law at node C
30 V 20 V V 0 50 3V
+

=
=0
20
20
20
20
V = 16.7 V
When solving for voltages or currents in a circuit using the loop method or the
nodal method, it is important to note that the number of independent equations needed is
equal to the number of unknowns in the circuit problem. With an equal number of
equations and unknowns, large numbers of unknowns can be solved for using a matrix
and linear algebra.
2.1.5 Attenuators

When we amplify the 1 mV electrocardiogram signal from the heart, the signal may be
undesirably reduced (attenuated) by the input resistance of the amplifier. For example,
Figure 2.8 shows that the resistance of the skin, Rs, may be 100 k and the input
resistance, Ri (input resistance of the oscilloscope used as an amplifier), is 1 M.

12

Chapter No. Title

Rs

1 mV

100 k

Electrocardiogram

Vo
1 M

Ri

Figure 2.8 The 1 mV signal from the electrocardiogram is attenuated by the resistive divider
formed by the 100 k skin resistance and the 1 M input resistance of the oscilloscope.

The current through the resistive divider is

i=

V
1 mV
=
= 0.91 nA
R 100 k + 1 M

The output voltage Vo is

Vo = iRi = (0.91 nA )(1 M ) = 0.91 mV

Thus the input voltage (ECG) has been undesirably attenuated by an amount that we can
calculate from the voltage divider equation

Vo
Ri
=
Vi Rs + Ri

(2.15)

Vo
Ri
1 M
=
=
= 0.91
Vi Rs + Ri 100 k + 1 M
Figure 2.8 is a typical example of voltage divider. With two resistors in series,
the output voltage is part of the input voltage. With adjusting the values of Ri and Rs, one
can flexibly obtain any percentage of the input voltage.
A potentiometer is a three-terminal resistor with an adjustable sliding contact
that functions as an adjustable voltage divider or attenuator. Figure 2.9 shows that if the
slider is at the top, vo = vi. If the slider is at the bottom, vo = 0. If the slider is in the
middle, vo = 0.5vo. Potentiometers are usually circular with a rotatable shaft that can be
turned by hand or a screwdriver. The potentiometer is useful to provide a variable gain
for an amplifier or for the volume control on a radio. Alternatively, a potentiometer can
be used as a two-terminal variable resistor by using the variable resistance between the
slider and only one end.

Section No. Title

vi

13

Slider
vo

Figure 2.9 A potentiometer is a three-terminal resistor with an adjustable sliding contact shown by
the arrow. The input signal vi is attenuated by the potentiometer to yield an adjustable smaller
voltage vo.

2.1.6 Common electrical instruments

Galvanometer

The galvanometer is a main component that is used in creating ammeters (devices that
measure current in a circuit) and voltmeters (devices that measure potential difference in
a circuit), as the galvanometer can be used in conjunction with other circuit elements to
measure current or potential differences of an electronic signal.
A common type of galvanometer consists of a coil of wire mounted in such a
way that it is free to deflect or rotate in the presence of a magnetic field provided by a
permanent magnet. The main principal behind the galvanometer makes use of a torque
that acts on the loop of current when a magnetic field is applied. The torque is
proportional to the amount of current that passes through the galvanometer, such that the
larger the current, the greater amount of deflection or rotation of the coiled wire.
Typically, an off-the-shelf galvanometer is not directly used as an ammeter,
since it has a large resistance (about 60 ) that would considerably reduce the amount of
current in the circuit in which the galvanometer is placed. Also, the fact that the
galvanometer gives a full-scale deflection for low currents (1 mA or less) makes it
unusable for currents greater in magnitude. If a galvanometer is placed in parallel with a
shunt resistor, Rp, with a relatively small resistance value compared with that of the
galvanometer, the device can be used effectively as an ammeter. Most of the current that
is measured will pass through this resistor (Figure 2.10(a)).
If an external resistor, Rs, is placed in series with the galvanometer, such that its
resistance value is relatively larger than that of the galvanometer, the device can be used
as a voltmeter. This ensures that the potential drop across the galvanometer doesnt alter
the voltage in the circuit to which it is connected (Figure 2.10(b)).
Older ammeters and voltmeters that used galvanometers with moving coils have
been largely replaced by digital multimeters with no moving parts.

14

Chapter No. Title

Galvanometer

Galvanometer

Rp
(a)

Rs
(b)

Figure 2.10 (a) When a shunt resistor, Rp, is placed in parallel with a galvanometer, the device can
be used as an ammeter. (b) When a resistor, Rs, is connected in series with the galvanometer, it can
be used as a voltmeter.

Wheatstone bridge

Often in biomedical instruments, unknown resistances or changes in resistance are

measured within a circuit. Many times, an electric element known as Wheatstone bridge
is used to measure these unknown resistances. The Wheatstone bridge has several
applications and is widely used in electronics.
One application of the Wheatstone bridge consists of a common source of
electric current (such as a battery) and a galvanometer (or other device that measures
current) that connects two parallel branches, containing four resistors, three of which are
known. One parallel branch contains one known resistance (R1) and an unknown
resistance (Rx) as shown in Figure 2.11. The other parallel branch contains resistors of
known resistances, R2 and R3.

Section No. Title

I1
+

I2
R1

R1

15

R3

Rx

Figure 2.11 A circuit diagram for a Wheatstone bridge. The circuit is often used to measure an
unknown resistance Rx, when the three other resistances are known. When the bridge is balanced,
no current passes from node a to node b.

In order to determine the resistance of the unknown resistor, the resistances of

the other three are adjusted and balanced until the current passing through the decreases
to zero. Since the voltage potentials at a and b are equal, the potential difference across
R2 must equal R1. Similarly, the potential difference across R3 must also equal Rx. With
this we have
I1R1 = I 2 R2

(1)

I1R3 = I 2 Rx

(2)

Dividing (2) by (1), we can eliminate the current, and solve for Rx
Rx =

R2 R3
R1

(2.16)

From Eq. (2.16), we can calculate the value of the unknown resistor.
The fact that a Wheatstone bridge is valuable for measuring small changes of a
resistance makes it also suitable to measure the resistance change in a device called a
strain gage. This device, commonly used in biomedical instruments to measure
experimental stresses, often consists of a thin wire matrix attached to a flexible plastic
backing and glued to the stretched metal. Stresses are measured by detecting changes in
resistance of the strain gage as it bends. The resistance measurement is made with the
strain gage as one or more elements in the Wheatstone bridge. Strain gages diffused into

16

Chapter No. Title

the diaphragm of an silicon integrated circuit blood pressure sensor are formed into a
Wheatstone bridge.
2.1.7 Capacitors

From Ohms law we know the voltage across a resistor is related to the current through
the resistor by the resistance value. Now we investigate two elements with different
relationships between voltage and current.
Conventionally, we use capital letters to denote variables that do not change
with time and small letters to denote variables that change with time.
A capacitor is a two terminal element in which the current, i, is proportional to
the change of voltage with respect to time, dv/dt, across the element (Figure 2.12), or
i=C

dv
dt

(2.19)

where C is capacitance and is measured in farads (F). One farad is quite large so in
practice we often see F (106 F), nF (109 F), and pF (1012 F). Common capacitor
values are 10 pF to 1 F. Capacitors are commonly used in a variety of electric circuits
and are a main component of electronic filtering systems (introduced in 2.3) in
biomedical instruments.

i
i

1
dv/dt

(b)

(a)

Figure 2.12 (a) Capacitor current changes as the derivative of the voltage. (b) Symbol of the
capacitor.

We can also represent v as a function of i and t

v(t ) = v(t0 ) +

1 t
idt
Ct
0

(2.20)

Section No. Title

17

where t0 and t are the beginning and ending times over which we observe the current, v(t0)
is the initial voltage across the capacitor. We usually choose a t0 of zero.
A capacitor usually consists of two conductors that are separated by an insulator.
The capacitance of a device depends on its geometry and the dielectric material. A
parallel-plate capacitor consists of two parallel plates of equal area A, separated by a
distance d (Figure 2.13). One plate has a charge +Q and the other Q, where the charge
per unit area on either plate is = Q/A. If the charged plates are very close together,
compared to their length and width, we can assume that the electric field is uniform
between the plates and zero everywhere else.
+Q

Area = A

Figure 2.13 Diagram of a parallel plate capacitor. The component consists of two parallel plates of
area A separated by a distance d. When charged, the plates carry equal charges of opposite sign.

We can then see that the electric field between the plates is
E = o = Q o A

(2.21)

where 0 is the permittivity of free space (8.85 pF/m)). Since the potential difference
between the plates is equal to Ed, we find that the capacitance is
C = 0 A d

(2.22)

Capacitances can be found in many different configurations within a circuit.

Two capacitances, or several capacitances, can be reduced to one capacitance value by
simple algebraic expressions depending on the situation. For the series capacitances in
Figure 2.14(a), the equivalent capacitance, Ce, can be calculated by

18

Chapter No. Title

1

1
CC
1
= 1 2
Ce =
+

C1 + C2
C1 C 2
1
1
1
1
=
+
+
+ mm
Ce C1 C2 C3

2 capacitors
(2.23)
multiple capacitors

For the parallel capacitances in Figure 2.14(b), the equivalent capacitance, Ce, can be
calculated by
2 capacitors
Ce = C1 + C2
Ce = C1 + C2 + C3 + mm multiple capacitors

(2.24)

C1

C1

C2

C2

(a)

(b)

Figure 2.14 (a) A series combination of two capacitors. (b) A parallel combination of two
capacitors

2.1.8 Inductors

An inductor is a two terminal element in which the voltage, v, is proportional to the

change of current with respect to time, di dt , across the element (Figure 2.15), or
v=L

di
dt

(2.25)

where L is the inductance. Its unit is the henry (H). Like the farad, this unit is large so we
use more convenient units such as mH (103 H) and H (106 H). This is usually taken to
be the defining equation for the inductance of any inductor, regardless of its shape, size
or material characteristics. Just as resistance is a measure of opposition to a current,
inductance is a measure of the opposition to any change in current. The inductance of an
element often depends on its geometry.

Section No. Title

19

V
L

1
di/dt

(b)

(a)

Figure 2.15 (a) Inductor voltage changes as the derivative of the current. (b) Symbol of the
inductor.

We can also represent i in terms of v and t

i (t ) = i (t0 ) +

1 t
vdt
Lt

(2.26)

+
? 0e?

L=2H

Figure 2.16 Simple inductor circuit.

Example 2.6 In Figure 2.16, an inductor is connected, at time t = 0 by a switch, to a

voltage source whose signal can be described as 10e5t V. Derive an equation that
describes the current through the inductor as a function of time.
Using Eq. (2.26) we can solve for i(t). There is no current through the inductor
at t = 0 so i(t0) = 0.

i (t ) = i (t0 ) +

1 t
vdt
Lt
0

1 t
1 10 5t
i (t ) = 0 +
10e5t V dt =
e

2H0
2 5
i (t ) = e 5t A

20

Chapter No. Title

For multiple inductors in series, the equivalent inductance Le, can be calculated
by
Le = L1 + L2 + L3 + mm

(2.27)

For multiple inductors in parallel, Le can be calculated by

1
1
1
1
=
+
+
+ ll
Le L1 L2 L3

(2.28)

2.1.9 First-order systems

Consider the simple resistor and capacitor (RC) circuit shown in Figure 2.17(a).
Suppose that at time t = 0 the voltage across the capacitor is vC(0). For vC(t)
when t 0, we have
VC(t)
1
+

+
R

iR

VC(0)

iC

(a)

(b)

Figure 2.17 (a) Simple RC circuit with v(0) on capacitor at time t = 0. (b) Normalized voltage
across the capacitor for t 0 (normalized means the largest value is 1).

vR = vC and iR + iC = 0

vC
dv
+C C =0
R
dt

(2.29)

Upon rearranging, we get

dvC
1
+
v =0
dt
RC C

(2.30)

This equation contains the variable vC and its first derivative dvC dt . This is a
first-order differential equation. Many complex systems can be reduced to multiple firstorder systems, which are more easily to be solved and the high order systems.

Section No. Title

21

The solution of this differential equation, assuming at t = 0, vC (t) = vC (0), is

vC (t ) = vC (0)e

t
RC

(2.31)

Figure 2.17(b) shows vC(t) versus t for t 0. Also, we have

t

iC (t ) = iR (t ) =

vC (t )
v (0)
= C
e RC
R
R

(2.32)

From these results we can see that voltage and current in the circuit decrease as
time increases and approach zero as time goes to infinity. The voltage and current in this
circuit can be described by equations of the form

f (t ) = Ke

(2.33)

where K and are constants. The constant is called the time constant of the first-order
system. In this case, = RC. When time t reaches , the amplitude of vC (t = ) is vC(0)E
1.

When t = 3, the value of vC(t) is down to only 5% of the value of vC(0). We can
assume the value of vC(t) is practically zero after five time constants.
In the circuit shown in Figure 2.17(a), there was no independent source (or
input), only an initial state of the circuit. For this reason, we call the voltage and current
we obtained the natural response.

Example 2.7 Consider the circuit in Figure 2.17(a). Suppose R = 1 k, C = 40 F and

the initial voltage across the capacitor at t = 0 is 10 V. How long will it take for the
capacitor to discharge to half of its original voltage?
This simple RC circuit can be solved with a first-order differential equation.
vC(0) = 10 V, so half of that voltage would give us vC(t) = 5 V. Using Eq (2.31) and
algebra we can solve for t.

vC (t ) = vC (0)e RC
v (t )
t
ln C =
RC
vC (0)
vC (t )
= (1 k )(40 F)ln (0.5) = 27.7 ms
t = RC ln

vC (0)
Now let us consider the case of nonzero input using the circuit shown in Figure
2.18(a). From t = to just before t = 0 (t = 0), the switch has been open. Therefore, no

22

Chapter No. Title

current has been flowing and no charge is in the capacitor. At t = 0, we close the switch
which causes current to flow and charge to accumulate in the capacitor.
+

vR

R
C

VC(t)
V

I
vC

VC(t)

(a)

(b)

Figure 2.18 (a) Series RC circuit with voltage step input at time 0. (b) Normalized voltage across
the capacitor.

Using KVL, the equation for this current is

V =C

dvC vC
+
dt
R

for t 0

or
dvC
1
V
+
v =
dt
RC C RC

(2.34)
for t 0

The solution of the equation is

vC = V Ve

t
RC

The voltage across the resistor, vR, and current, i, are

(2.35)

Section No. Title

t
RC
t

e RC

vR = Ve
i=

V
R

23

(2.36)

The voltage vC(t) is shown in Figure 2.18(b). The voltage across the resistor
decreases exponentially and the voltage across the capacitor increases exponentially.
From the definition of a capacitor, we know the voltage across it cannot change abruptly
or the current will become infinite. When the switch is closed, the voltage on the
capacitor is still zero and the voltage drop across the resistor is V, the source voltage. As
time increases, the capacitor becomes charged and as time approaches infinity, all the
source voltage is across the capacitor while the voltage drop across the resistor is zero.
Example 2.8 Consider the circuit in Figure 2.18(a). Suppose that V = 12 V, R = 1 k and
C = 10 F. At time t = 0 the switch closes. Sketch the voltage across the capacitor versus
time. How long will it take for the capacitor to become 90% of the voltage source?
Using Eq (2.35):

vC = V Ve

vC = 12 12e

t
RC

t
0.01

90% of the voltage source is 10.8 V. Substituting this value in for vC in the
above equation and solving for t yields a time of 23 ms with a time constant of 10 ms.

vC (V)

12
10.8

0.023
Time (s)
Figure 2.19 Plot of vC for Example 2.8.

0.05

24

L/R.

Chapter No. Title

We can also use a resistor and an inductor to make a first-order circuit where =

2.1.10 Frequency

Sinusoidal waves are widely used in electrical circuits. Figure 2.20 shows two sinusoidal
waveform, which can be represented as
A sin( 2ft + ) = A sin(t + ) = A sin(
where

2
t +)
T

(2.37)

A = amplitude of sine wave

f = frequency of sine wave in hertz (Hz)
= 2f = angular frequency of sine wave (radians per second)
= phase angle of sine wave (radians)
T = 1/f = period (seconds)

Amplitude

sine

0
T

cosine

Figure 2.20 One period, T, of the sine and cosine waveforms.

Note the sine function lags 90 in phase behind the cosine function.
Example 2.9 Figures 2.21 and 2.22 show a comparison of a sinusoidal waveform with
different frequency and different phase angle.
In Figure 2.21 the solid line shows the function y = sin(t), the dashed line
shows the function with the frequency doubled, y = sin(2t). By doubling the frequency
the period is reduced by half by T = 1/f.

Section No. Title

25

v (V)

0.5
0
0.5
1

3
t (s)

Figure 2.21 Sinusoidal waveforms with different frequencies.

In Figure 2.22 the solid line again shows the function y = sin(t), the dashed
line shows the function with a phase shift of 180 ( in radians), y = sin(t ). The
minus sign in front of the phase angle shifts this waveform to the right and can be said
that it is leading the original waveform. Also with a phase shift of 180, the two
waveforms are said to be completely out of phase.

v (V)

0.5
0
0.5
1

3
t (s)

26

Chapter No. Title

Figure 2.22 Sinusoidal waveforms with 0 phase angle (solid) and 180 phase angle (dashed).

When a voltage or current source is a sinusoidal wave, Kirchhoffs laws still

apply in differential form to include capacitors and inductors. Since sinusoidal sources
change with time and its differential is still sinusoidal, a more efficient way to solve the
equations is to use complex sinusoidal forms.
For example, we can write Acos(t + ) in complex sinusoidal form as
Ae j (t + ) = Ae j e jt

(2.38)

This representation comes from Euler's identity and has the same amplitude, frequency,
and phase angle as the noncomplex form. Let us denote the complex number

by

A = Ae j

(2.39)

A = A

(2.40)

which is known as a phasor, with bold type meaning it is a vector. Equation (2.40) can be
read as, "The vector A has a magnitude of A and an angle of ."
Let us consider a resistor. If the current and voltage of the resistor are all
complex sinusoids, say
i = Ie j ( t + )
v = Ve j ( t + )

(2.41)

v = Ri
Ve j ( t + ) = RIe j ( t + )
Ve j = RIe j or V = RI

(2.42)

Then by Ohm's law

Eq. (2.42) shows that = , which means the voltage and the current are in
phase across the resistor (Figure 2.23(a)). We can re-write this as the phasor
V = RI

(2.43)

As to an inductor, the voltage across it is given by

v=L
The V-I relationship can be written as

di
dt

(2.44)

Section No. Title

27

Ve j = jLIe j or V = jLI

(2.45)

V = jLI

(2.46)

Therefore

and = + 90, which shows that current lags the voltage by 90 (Figure 2.23(b)).
Using the methods presented in Eq. (2.44) through (2.46) yields
V=

1
I
jC

(2.47)

and = 90, which shows that current leads the voltage by 90 (Figure 2.23(c)).
Imax

iR

Imax

Vmax

vR

Vmax

iC

Imax
vL

Vmax

vC

iL
(a)

(c)

(b)

Figure 2.23 Plots of the current and voltage as a function of time. (a) With a resistor both the
current and the voltage vary as sin(t), the current is in phase with the voltage, meaning that when
the current is at a maximum, the voltage is also. (b) For an inductor, the current lags behind the
voltage 90. (c) For a capacitor, the current leads the voltage by 90.

From the equations above, we can see that there is a general equation in phasor
notation for the resistor, capacitor, and inductor. That equation is
V = ZI

(2.48)

where Z is the impedance of the element. For the three elements,

ZR = R

ZC =

1
jC

Z L = jL

(2.49)

With impedance form of Eq. (2.49) and the method introduced in section 2.1.4,
the relationship of voltage and current is expressed in algebraic form instead of
differential form. Phasor equations simplify the circuit analysis and can be used by
computers to solve the problem.
Note that the impedance of capacitors and inductors changes with frequency.
Therefore, the performance of the circuit will also change with frequency. This change in
circuit performance is called the frequency response of the circuit.

28

Chapter No. Title

Example 2.10 The current through a 50 mH inductor is 1000 mA. If = 1000 rad/s,
what is the voltage across the inductor?
Using the phasor equation in Eq. (2.46), we can solve for V.
V = jLI
V = j1000(0.050 )(1000 mA ) = 590 V
2.1.11 Series and parallel impedances
Now that the impedances for the three passive circuit elements are defined (Eq. 2.49), it
is time to study two types of configurations for them in a circuit. Figure 2.24(a) shows
two impedances in series. These two impedances, and in general infinitely many of them,
can be combined into one equivalent impedance. For the series impedances in Figure
2.24(a), the equivalent impedance, Ze, can be calculated by
Z e = Z1 + Z 2

(2.50a)

For multiple impedances in series, the equivalent impedance Ze is

Z e = Z1 + Z 2 + Z 3 + 

(2.50b)

For the parallel impedances in Figure 2.24(b), the equivalent impedance, Ze, can be
calculated by

Ze =

ZZ
1
= 1 2
1
1
Z
1 + Z2
+
Z1 Z 2

(2.51a)

For multiple impedances in parallel, the equivalent impedance Ze is

1
1
1
1
=
+
+
+ 
Z e Z1 Z 2 Z3

(2.51b)

A shorthand notation for indicating parallel circuit elements as in Eq. (2.51) is to use a
double vertical bar between the parallel elements, such as Ze = Z1||Z2.
Using these definitions, it is possible to reduce large complicated circuits down
to (ideally) one or two impedances and a voltage source. Figure 2.24(c) shows the
equivalent impedance of the combined series or combined parallel circuit.

Section No. Title

29

Z1

Z2

Z1

Z2

(a)

(b)

Ze

(c)

Figure 2.24 (a) Series circuit. (b) Parallel circuit. (c) Single impedance equivalent.

2.1.12 Electrical safety

Because safety is paramount in considering biomedical devices in the real world setting,
great emphasis is placed on currents and voltages that exist in an apparatus and what
dangers these may pose to a patient or operator who might use the device.
Electric shock may result in fatal burns, or can cause muscles or even the heart
to malfunction. Often the degree of damage depends on the magnitude of current that is
applied, how long the current acts, and through which point on the body the current
passes. Typically, skin currents of 5 mA or less may cause a light sensation of shock, but
usually do no damage. Currents larger than 10 mA tend to cause muscles to contract,
making the person unable to let go of a live wire. If currents of 100 mA pass through the
body, even for a few seconds, respiratory muscles can become paralyzed and breathing
stops. The heart can go into ventricular fibrillation, which is fatal (macroshock). If an
electrode or catheter is within the heart, current must not exceed 10 A because the
current density is high (microshock). Contact with live wires or voltages above 24 V is
not recommended.
Often in biomedical engineering, currents and voltage outputs must be amplified,
attenuated, or filtered safely in order to achieve the best possible electronic signal with
minimal harm to the patient or operator. In the following sections, we will explore ways
in which this can be done.

2.2 Amplifiers
Most bioelectric signals have a very small magnitude (on the order of millivolts or
microvolts) and therefore require amplification so that users can process them. This
section will emphasize the operational amplifier (op amp) for use in amplifier design.

30

Chapter No. Title

2.2.1 Basic op amp

An op amp is a high-gain dc differential amplifier (differential meaning that any signals

not the same at both inputs are greatly amplified). An op amp is made of many transistors
and other components that are integrated into a single chip. We need not consider the
components within the chip, but simply the terminal characteristics. Focusing on the
terminal behavior of op amps allows us to see many applications for this circuit in
bioinstrumentation.
Figure 2.25(a) shows the equivalent circuit of an op amp. Figure 2.25(b) shows
the symbol of the op amp. v1 and v2 are the two input terminals, vo is the output terminal
and V+ and V are the positive and negative power supplies. The gain (A in Figure 2.25)
is defined as an increase in voltage by an amplifier, expressed as the ratio of output to
input. Gain is also called amplification
RS
v1

vo

Rd

A(v2 v1)

v1

v2

V+
A

vo

v2
(a)

(b)

Figure 2.25 (a) Equivalent circuit for op amp. (b) Symbol of op amp. Many times V+ and V are
omitted in the op amp symbol, but it is understood that they are present.

2.2.2 Ideal op amp

When designing a circuit, we assume that we are using an ideal op amp. Then we
investigate whether or not the nonideal characteristics affect the performance. If they do,
we revise the initial design to accommodate the nonideal characteristics.
For an ideal op amp, we make the following assumptions
1. A = (gain, or amplification of the input, is infinity)
2. vo = 0 when v1 = v2 (no offset voltage)
3. Rd = (differential input impedance is infinity)
4. Rs = 0 (output impedance is zero)
5. Bandwidth = . (Bandwidth will be discussed in greater detail in sections
2.27 and 2.3.)
Two basic rules should be stated before we discuss designing circuits with op amps.

Section No. Title

31

Rule 1: The two input terminals are at the same voltage.

If the two input terminals were not at the same voltage, the differential input voltage
would be amplified by infinite gain and yield infinite output voltage. This is not realistic.
For an op amp with a power supply of 15 V, most manufactures guarantee a linear
output range of 10 V. Also the op amp saturates about 13 V, which means that when
vo exceeds 13 V, it saturates and further increases in vi will not produce change in the
output.
Rule 2: No current flows into either of the two input terminals.
This is true because we assume the input impedance of an op amp is infinity and no
current flows into an infinite impedance.
2.2.3 Inverter

Figure 2.26 shows an inverting amplifier, or inverter. It is the most widely used circuit in
instrumentation. Note that Rf is connected from the output to the negative input terminal
and thus provides negative feedback. The negative feedback increases the bandwidth and
lowers the output impedance of the op amp circuit.
Rf

i
i

vi
Ri

A
+

vo

Figure 2.26 An inverting amplifier. The gain of the circuit is Rf/Ri.

In Figure 2.26, the positive input terminal of the op amp is connected to ground
and is always at 0 V. By rule 1, the negative input of the op amp is also at 0 V and
remains at 0 V no matter what happens to the other components of the circuit. This
clamping of the negative input to 0 V is called a virtual ground.
By rule 2, we know no current flows into the op amp. Therefore, the current
flows through Ri and then through Rf to the output terminal, then through the output
circuit of the op amp to ground.
Using KCL at the terminal
iin = iout

vi 0 0 vo
=
Ri
Rf

vo =

v
R
Rf
vi o = f
vi
Ri
Ri

(2.52)

32

Chapter No. Title

The output vo has the opposite sign of the input vi and is amplified by an amount Rf Ri .
When using op amps in circuits it is important to consider the output impedance
of any source that they are connected to, as this value will affect the total input resistance
and the gain of the op amp.
Rf
Rs

vi

Ri

vs= 1 V

A
+

vo

Figure 2.27 Inverter circuit attached to a generator that contains an internal resistance.

Example 2.11 Suppose in Figure 2.27 that vi = 1 V, Rf = 4 k and Ri = 1 k, we would

have, from Eq. (2.52), a gain of 4 and therefore, an output voltage of 4 V. However if
the signal generator shown in Figure 2.27 were connected and had output impedance Rs,
this impedance would add to Ri of the op amp and the new input impedance would be Rs
+ Ri. Rs would therefore change vo as well as the gain of the circuit. If Rs = 150 , then
using Ohms law we obtain the gain

vo
Rf
=
= 3.48
vs
Ri + Rs
which is much less than our desired value.
2.2.4 Noninverting amplifier

Figure 2.28(a) shows the circuit for a noninverting amplifier. From rule 1, we know the
voltage of the negative input always equals the voltage of the positive input terminal,
which in this case is vi. From Figure 2.28(a) we have

vi 0 vo vi
=
Ri
Rf

vo Rf + Ri
R + Ri
R
=
or vo = f
vi = vi 1 + f
vi
Ri
Ri
Ri

(2.53)

Section No. Title

Ri

Rf

A
+

vi
(a)

33

vo
vi

vo

+
(b)

Figure 2.28 (a) A noninverting amplifier has a gain of (Rf + Ri)/Ri. (b) A follower, or buffer, with
unity gain.

Notice the output vo has the same sign as the input voltage vi. The amplifier gain
is determined by the ratio (Rf + Ri)/Ri.
If we choose Ri = (open circuit), the circuit can be simplified as Figure
2.28(b). The gain, (Rf + Ri)/Ri, would be unity. Therefore vo = vi, or the output voltage
follows the input voltage. At first glance, this circuit does nothing, since the output equals
the input. However, this circuit is very useful as a buffer, to prevent a high source
resistance from being loaded by a low-resistance load.
Loading can be expressed by the voltage divider equation (Eq. 2.15). In Figure
2.8, if Ri is a significant fraction of Rs, we would get an attenuated signal for Vo. We say
that the source resistance, Rs, is being loaded by Ri. This can be seen mathematically in
Eq. 2.15as Rs increases to infinity, Vo/Vi approaches zero. In general, we like to have
Ri >> Rs, which is what the buffer circuit does. It provides an extremely high input
impedance and an extremely low output impedance.
Example 2.12 Suppose the signal generator in Figure 2.27 were connected in a
noniverting configuration (Figure 2.29), will the generators output impedance affect the
gain?
No, since no current will flow into the positive input terminal, vi = 1 V and the
gain would remain (Rf + Ri)/Ri. If Rf = 3 k and Ri = 1 k, the gain would be 4,
independent of the value of Rs.

34

Chapter No. Title

Rf

Ri

Rs

vo

vi
vs= 1 V

Figure 2.29 The gain of the noninverting amplifier is not affected by the addition of the impedance
Rs due to the generator.

2.2.5 Differential amplifiers

Figure 2.30 is a simple one-op-amp differential amplifier. Current flows from v2 through
R1 and R2 to the ground and no current flows into the op amp. Thus, R1 and R2 act as a
voltage divider.
R1

R2

v1
R1

v3

v3

v2

vo

R2

Figure 2.30 A differential amplifier uses two active inputs and a common connection.

The voltage that appears at the positive input terminal determines the voltage at
the negative terminal. The top part of the amplifier works like an inverter.
The voltage at the positive input terminal is
v3 =

R2
v
R1 + R2 2

(2.54)

Section No. Title

35

Since there is no flow of current into the op amp,

i=

v1 v3 v3 vo
=
R1
R2

(2.55)

Combining these two equations yields

vo =

R2
(v v )
R1 2 1

(2.56)

This is the equation for a differential amplifier. If the two inputs are hooked
together and driven by a common source, then the common-mode voltage is equal to Vc
(v1 = v2). The differential amplifier circuit gives an output of 0 and the differential
amplifier common-mode gain (Gc) is 0. When v1 v2, the differential voltage (v2 v1)
produces a differential gain (Gd), which equals R2/R1.
An important application of differential amplifier is to remove interference from
the input signal. Take ECG for example, the wires from patient to ECG machine are
about 1 ~ 2 m in length and are exposed to different interference such as power line
interference. As there wires have the almost same amount of noise since they are exposed
to the same common noise sources, this common noise is rejected by the differential
amplifier according to Eq. (2.56). The output of the amplifier solely depends on the input
from the patient.
In some instances, the signal on both input lines (such ECG input terminals) is
corrupted by a surrounding interference (such as power line interference) and is not
acceptable for us to process. From Eq. (2.56) we know that only different signals input
into the differential amplifier will be amplified. Therefore, the common interference on
both inputs will not be amplified.
No differential amplifier perfectly rejects the common-mode voltage. To
quantify the performance of a differential amplifier, we use the term common-mode
rejection ratio (CMRR), which is defined as
CMRR =

Gd
Gc

(2.57)

In real applications a CMRR greater than 100 may be acceptable, depending on

the situation. A high-quality differential amplifier may have a CMRR greater than
1,000,000 (120 dB).
To measure CMRR, We have to know both Gd and Gc. We can apply the same
voltage to the two different inputs and measure the output to obtain Gc. Then we can
connect one input to ground and apply a voltage to the other one. From the output, we
can obtain Gd. With these two numbers, we can calculate the CMRR from Eq. (2.57).

36

Chapter No. Title

As with the inverting amplifier, we must be careful that resistances external to

the op amp circuit do not affect the circuit. A differential voltage connected to a
differential amplifier can have two output impedances affecting the op amp input
impedance. This is especially a problem when the added impedance fluctuates (for
example, skin resistance fluctuates). One solution to the problem is to use buffer
amplifiers as in Figure 2.31. Because no current flows into the input terminal, there is no
current across the resistor. Thus, there is no loss of voltage.

Rs

R2

vd/2

+
_
vd/2
_

R1

Rs

R1

vo

vc

R2

Figure 2.31 Differential amplifier attached to a differential voltage that contains varying
impedances. Adding buffers ensure that fluctuations in Rs do not affect the gain.

R4 = 1 k

R5 =1.5 k

4V
R1 = 1 k

v3
v3

8V
R2 = 500

A
+

vo

R3 = 1 k

Figure 2.32 Differential amplifier for Example 2.13

Example 2.13 For the differential amplifier in Figure 2.32, solve for vo.
Using KCL, we can solve for vo. First, the voltage at the positive terminal of the
amplifier is part of a voltage divider.

v3 =

R3
1 k
8V =
8V =6V
(R1 || R2 ) + R3
(1 k || 500 ) + 1 k

Section No. Title

37

Then because there is no flow of current into the op amp, using Kirchhoffs
current law, we can solve for vo.
4 V v3 v3 vo
=
R4
R5
4 V 6 V 6 V vo
=
1 k
1.5 k
vo = 9 V

i=

2.2.6 Comparators

A comparator is a circuit that compares the input voltage with some reference voltage.
The voltage output flips from one saturation limit to the other as the negative input of the
op amp passes through 0 V. If the input is greater than the reference, the output is the
negative saturation voltage. Otherwise, the output is the positive saturation voltage. Note
that the saturation voltage is always smaller than the voltage of the power supply for the
op amp.
The simplest comparator is the op amp itself. If a reference voltage is connected
to the positive input and vi is connected to the negative input, the circuit is complete.
However the circuit in Figure 2.33 is more widely used.
R1

Saturated
voltage

vi

vo
vref = vi

vref

R1
(a)

vi

vo

(b)

Figure 2.33 (a) A comparator. (b) The inputoutput characteristic of the comparator in (a).

The positive input is always at 0 V since it is grounded. Because the current

flows from vi to vref through two equal resistors, the voltage at the negative input is their
average value (vi + vref ) 2 . If vi > vref, the voltage at the negative input is larger then
the voltage at the positive input, which is 0 V. Because the op amp is open loop and its
gain is infinite, the output voltage vo saturates at the negative saturation value, which is
usually about 13 V. On the other hand, if vi < vref, the output is the positive saturation
value, which is about 13 V.
An application of the comparator is for detecting heartbeats from the ECG signal
as shown in Figure 2.34. The normal ECG signal contains a P wave, a QRS complex, and

38

Chapter No. Title

a T wave. The QRS complex normally has a much larger amplitude than the other two
waves. A threshold is set, which is larger than the P and T waves but smaller than the R
waves. Every time a QRS complex passes through the comparator, the comparator gives
out a pulse, which corresponds to the heart beat. These pulses are sent to a counter and
the heart rate can be calculated.

R
Threshold
T
P
Q

Figure 2.34 Heart beat detector uses a comparator to determine when the R wave exceeds a
threshold.

2.2.7 Frequency response

An ideal op amp has unlimited bandwidth; its gain is same for low frequency and high
frequency. However, an op amp is made of several stages of circuitry to obtain very high
gain. Each of the stages has some stray capacitance or junction capacitance associated
with it. This capacitance reduces the high frequency gain of the op amp. Figure 2.35 is
the typical frequency response of an open loop op amp.
It would appear that an open-loop op amp has very poor frequency response,
since its gain is reduced when the frequency exceeds 10 Hz. Because of this, we never
use an open loop op amp to build an amplifier. Instead, we introduce large negative
feedback to control the gain of the amplifier circuit. For example, if we have an op amp
with a circuit gain of 10 (Figure 2.35), the frequency response is flat up to 100 kHz and is
reduced above that frequency.

Section No. Title

Gain

Circuit bandwidth

Ideal
gain
Typical open
loop gain
Circuit gain of 10

10
1

39

100

102
104
Frequency (Hz)

106

Figure 2.35 The typical op amp open loop gain is much larger, but less constant, than the circuit
gain. However the in circuit bandwidth is larger than the open loop bandwidth.

We use the unity gainbandwidth product to quantify the frequency performance

of an op amp. The gainbandwidth product is the product of gain and bandwidth at a
particular frequency. Figure 2.35 shows that the unity gain bandwidth is 1 MHz. Along
the entire curve with a slope of 1, the gainbandwidth product is constant at 1 MHz.
Thus, for any amplifier circuit, we can obtain the bandwidth by dividing the unity gain
bandwidth product by the amplifier circuit gain.
2.3 Filters

To obtain a specific output from a device, it is sometimes necessary to electronically sort

through and pick out exactly what is desired from the incoming mass of signals. Filters
can simplify this process by altering the amplitude and phase of the input signal in a
desired way. In most applications, the signal we obtain contains more frequency
components than we want, so we use a filter to reject undesired components. Take a
stereo for example; it has bass (low pass filter) and treble (high pass filter) controls to
adjust the pitch of the sound. It also has oscillator circuitry (band pass filter) to tune to
different radio stations. This is the basic idea behind electronic filters. More specifically,
a filter is an electronic circuit that, by design, attenuates sinusoidal voltage signals at
undesired frequencies and passes signals at preferred frequencies. In other words, it
rejects undesired components of a signal. Similarly, in the case of biomedical
applications, filters are useful when recording physiological signals such as the ECG
(electrocardiogram) and EEG (electroencephalograph).
2.3.1 Inputoutput relationship of filters

The changes made to a signal depend on the frequencies contained in the signal and the
design of the filter. Generally a filter is inserted between an input source and an output.
Suppose the input, vi, is a signal with a frequency of

40

Chapter No. Title

vi = Vm sin(t + )
vo = Vm T ( ) sin(t + + ( ))

(2.58)

where Vm is the amplitude of the input signal, |T( )| is the amplitude of the transfer
function, or filter, and () is the angle of T().
Thus the filter multiplies the amplitude of the input sine wave by |T( )| and adds
to the phase of the sine wave (). Since |T( )| and ( ) are functions of frequency, they
are named the amplitude response and phase response, respectively, of the filter.
For example, the ideal low-pass filter has a magnitude response of
1
T( f ) =
0

if f < f c
if f f c

(2.59)

where f = /2.
The low-pass filter passes all frequency components lower than fc and rejects all
frequency components higher than fc. Realistically, building a filter with such a sharp
transition is not possible since the transition from passband (T(f) = 1), to stopband (T(f) =
0) takes some finite number of frequencies.
Figure 2.36 shows the magnitude characteristics of four widely used filters: lowpass, high-pass, bandpass and bandstop filter. All these filters have a passband, where the
filter passes the frequency components, and a stopband, where the filter rejects or
attenuates the frequency components.

Section No. Title

T(f)

41

T(f)
PB

1.0

PB

1.0

0.1

0.1

0.01
1

0.01

100 f

fc 10

10 fc

(a)

T(f)

(b)

100 f

T(f)

PB

1.0

PB

1.0

0.1

PB

0.1

0.01
1

f1 10

f2

0.01

100 f

(c)

f1

10

f2

100 f

(d)

Figure 2.36 (a) Low-pass filter. (b) High-pass filter. (c) Bandpass filter. (d) Bandstop filter. (PB
denotes passband)

Circuits can be used to approximate ideal filters. The closer the actual circuit
characteristics approach the ideal response (i.e. the steeper the transition band), the better
the approximation. However, the complexity of the circuit increases as the ideal response
is more closely approximated.
2.3.2 Low-pass filter

Low-pass filters pass low-frequency signals and attenuate high-frequency signals. The
corner frequency, c, distinguishes the passband from the stopband.
The first-order RC and RL networks are approximations of a low-pass filter.
Figure 2.37 shows the RC and RL filter.

42

Chapter No. Title

vo

vi

(a)

+
vi

vo

(b)

Figure 2.37 Low-pass filter. (a) RC circuit. (b) RL circuit.

For the RC filter in Figure 2.37(a), the frequency response is calculated using
the voltage divider principle for complex impedance
T ( ) =

1/ jC
1
1
=
=
R + 1 / jC 1 + jRC 1 + j c

(2.60)

where

c =

1
RC

(2.61)

For the RL circuit in Figure 2.37(b), we have

T ( ) =

R
1
=
R + jL 1 + j c

(2.62)

where

c =

R
L

(2.63)

The magnitude and phase responses of these filters are similar to those in Figure
2.36(a). At = c (f = fC), the magnitude is 1 2 times the magnitude at = 0. We
designate c the corner frequency. This frequency is also known as the half power
frequency because at = c the power delivered to the load is half of what it is at = 0.
A typical use of a low-pass filter in bioinstrumentation is to remove the radio
frequency (RF) interference from an ECG signal. A combination of unshielded ECG
electrode leads and long electrode-to-amplifier connecting wires causes RF signals from
other medical equipments to couple to the amplifier. The RF signal is rectified by the
transistor junctions and causes interference below 150 Hz. However, frequency
components of the ECG also lie below 150 Hz. Therefore, the ECG is always sent

Section No. Title

43

through a low-pass filter before it is amplified in order to filter out radio-frequency

signals. For example, a 10 k resistor and 300 pF capacitor can make the circuit in
Figure 2.37(a), which will minimize RF interference.
2.3.3 High-pass filter

High-pass filters also be can realized with first-order RC or RL circuits (Figure 2.38).
R

+
vi

vi

vo

(a)

vo

(b)

Figure 2.38 High pass filter. (a) RC circuit. (b) RL circuit.

The frequency responses of these two filters have a similar form:

T ( ) =

1
1 j c

(2.64)

where
1
RC
R
c =
L

c =

for the RC circuit

(2.65)

for the RL circuit

The frequency response of the first order high-pass filter is similar to that shown
in Figure 2.36(b). The corner frequency of the high pass filter is c.
The ECG signal always has some dc offset and low-frequency artifacts due to
the electrodes. A high-pass filter with a corner frequency of 0.05 Hz is used to filter the
signal. A resistor of 3.2 M and a capacitor of 1 F can be used to make a high-pass
filter that passes frequencies only higher than 0.05 Hz.
2.3.4 Other filters

More complex circuits, such as second, third, or higher order circuits, can be used to
make the changes in magnitude from passband to stopband more abrupt.

44

Chapter No. Title

Second order circuits, such as RLC circuits, can be used to realize bandpass and
bandstop filters. Or a combination of a low-pass circuit and a high-pass circuit can be
used together to achieve the same filtering effect.
For example, a device requires a bandpass filter with corner frequencies 1 and
2 ( 1 < 2). In this case, a low-pass filter with a corner frequency of 2 and a high-pass
filter with corner frequency of 1 are suitable. These two filters can be cascaded so the
signal will first go through the low-pass filter and then through the high-pass filter. The
high-pass filter rejects the frequency components smaller than 1 while the low-pass
filter rejects frequency components greater than 2. As a result, only the frequency
components between 1 and 2 will pass through the cascaded filter. This combination
of a high-pass and low-pass filter is used as an approximation of a bandpass filter (Figure
2.39).
Input signal

Low-pass filter
corner frequency
2

High-pass filter
corner frequency
1

Output

Figure 2.39 A low-pass filter and a high-pass filter are cascaded to make a bandpass filter.

Similarly, if a single filter cannot satisfy the requirement of signal processing,

several filters can be cascaded together to form a multi stage filter until they satisfy the
need for application. However, keep in mind that the response of cascaded filters is more
complex than that for a single filter.

2.4 Analog-to-digital (ADC) and digital-to-analog (DAC) conversion

The past sections discuss analog signals, which are usually continuous in amplitude,
phase, and time. Now we will discuss digital circuits, in which signals take on a limited
number of values over discrete time. The most common digital system employs two
number values and is defined as a binary system. In a binary system, voltages are either
high or low, and the symbols 1 and 0 are used to represent these two possible levels.
The simplest binary system is a switch to control light. When a switch is closed
and the light goes on, we denote that state as 1. When the switch is open and then the
light is turned off, the state is 0. There are many examples of binary systems.
The main advantage of digital signals is their high error tolerance. In digital
systems, we often define voltages in some range as high or 1 and in other ranges as low
or 0. For example, in TTL logic circuits, we define a voltage of 4 V to 5 V to be high and
0 V to 1 V to be low. The voltages between 1.0 V and 4.0 V are undefined. If a high state
(5 V) signal is corrupted to 4.5 V during transmission due to interference, it is still
assumed as high state at the receiver end and can be restored to 5.0 V. For an analog
signal, the corruption is more difficult to recover because the signal corruptions, such as
transmission loss, interference and signal deterioration, are random process.

Section No. Title

45

Digital circuits play a very important role in todays electronic systems. They
are employed in almost every aspect of electronics, including communications, control,
instrumentation, and computing. This widespread usage is due to the development of
inexpensive integrated circuit (IC) packages. The circuit complexity of a digital IC ranges
from a small number of transistors (~100 to 10,000) to a complete computer chip (42
million transistors for a Pentium 4 chip).
2.4.1 Number systems

The most commonly used numbering system is the decimal system, which was devised
by Hindu mathematicians in India around 400 A.D. Although other numbering systems
have been developed, the decimal system is the most natural one and is widely accepted.
A decimal number is represented by a string of digits. Depending on its position
in the string, each digit has an associated value of an integer raised to the power of 10.
For example, the decimal number of 215, can be computed as
215 = 2 102 + 1 101 + 5 100
Conventionally, we only write the digits and infer the corresponding powers of
10 from their positions. In general, a decimal number can be written as a string of digits
as
An An 1 An 2 ... A1 A0

(2.66)

Each coefficient Ai is one of the 10 digits (0, 1, 2, 3, 4, 5, 6, 7, 8, 9).

The decimal number system is base 10 because the coefficients are multiplied by
powers of 10 and the system uses 10 distinct digits. In general, a number in a base r
number system contains digits 0, 1, 2,,r 1 digits and is expressed as a power series in
r with the form
An r n + An 1r n 1 + An 2 r n 2 + ... + A1r1 + A0 r 0

(2.67)

( An An 1 An 2 ... A1 A0 )r

(2.68)

or in position form

For the number 314 in a base-5 number system, we have

(314)5 = 3 52 + 1 51 + 4 50 = 75 + 5 + 4 = (84)10
In addition to decimal, we use binary, octal and hexadecimal number systems
when dealing with computers. These are base-2, 8, and 16 number systems, respectively.

46

Chapter No. Title

The binary number system is a base-2 system that uses the digits 0 and 1. Two
numbers are chosen because they can represent the two states of the digital system (e.g.
voltage high and low, or switch closed and open). A binary number can be converted to a
decimal number using Eq. (2.67). For example
(100101) 2 = 1 25 + 0 24 + 0 23 + 1 22 + 0 21 + 1 20
= 32 + 4 + 1 = (37)10
To change a decimal number to a binary number, the method of dividing the
integer number by the base successively until 0 results is most often used. The
remainders at every division form the binary number. The remainder of the last division
is the most significant bit (MSB) and the remainder of the first division is the least
significant bit (LSB). For example

11 2 = 5 with remainder 1 (LSB)

52 = 2
1
22 =1
0
1 2 = 0
1 (MSB)
so
(11)10 = (1011)2
Binary digital impulses, alone, appear as long strings of ones and zeros, and
have no apparent meaning to a human observer. However, if a digital-to-analog converter
(DAC) is used to decode the binary digital signals, meaningful output appears. Examples
of this output might be a computerized voice, a picture, or a voltage to control
mechanical motion.
2.4.2 Digital-to-analog converters (DAC)

The digital representation of a continuous analog signal is discrete both in time and
amplitude. Often an analog signal must be sampled discretely with an analog-to-digital
converter (ADC). Digital information is stored in a memory device for real time or future
processing. A digital signal needs to be converted to an analog signal with a digital-toanalog converter (DAC), since many devices need an analog signal for controlling,
monitoring or feedback. For example, modems use ADCs and DACs in converting digital
information in computers to analog signals for phone lines and back again to
communicate with Internet.
The DAC changes digital input to analog output. Figure 2.40 shows the static
behavior of a 3-bit DAC. All the combinations of the digital input word are on the
horizontal axis, while the corresponding analog outputs are on the vertical axis. The
maximum analog output is 7/8 of the reference voltage Vref. For each digital input, there
is a unique analog output. The difference between the consecutive analog outputs is

Section No. Title

47

1
1
Vref = 3 Vref
8
2
We call this value the resolution of the DAC. In general, for an n-bit DAC or ADC, the
resolution is
resolution =

1
Vref
2n

(2.69)

7/8 Vref

Analog output

6/8 Vref
5/8 Vref
4/8 Vref
3/8 Vref
2/8 Vref
1/8 Vref
0V
000

001

010

011

100

101

110

111

Figure 2.40 The ideal static behavior of a 3-bit DAC. For each digital input, there is a unique
analog output.

Equation (2.69) shows that if an application calls for high resolution, the DAC
needs many bits. DACs are available in 4, 8, 10, 12, 16, 20 or more bits. The more bits a
DAC has, the more complex and expensive it is.
A variety of circuit configurations are suitable for a DAC. We present the
simplest one: a voltage scaling DAC. Figure 2.41 is a simple 3-bit voltage scaling DAC.

48

Chapter No. Title

The Vref in Figure 2.41 is the reference voltage of the DAC. The analog output is derived
from this reference. The accuracy and stability of the reference should be carefully
considered in the circuit design.
The reference voltage is divided by the series resistor network to obtain the
desired discrete voltages on each node. The digital inputs are decoded by the decoder and
the corresponding switch is closed. This voltage is then sent out as the analog output.
Voltage scaling is suited to IC technology because this technology is optimized
for making many copies of the same structure. The value of R may have an error up to
50%, but as long as the difference between two resistors is still below 1%, the discrete
output retains the accuracy of 1%.
The output of the DAC is discrete and not continuous. Interpolation should be
used to make the output voltage smoother. Interpolation could be implemented by an
electronic circuit with a voltage holder, linear interpolation, or low-pass filter.

Section No. Title

b2

b0

3-to 8 decoder

Vref

b1

49

R
R
Vo
R
R
R
R
R

Figure 2.41 A 3-bit voltage scaling DAC converter.

2.4.3 Analog-to-digital converters (ADC)

Many signals used in biomedical engineering such as voltage, temperature, pressure,

strain, flow and speed, will originally be in analog form. These signals need to be
converted to a digital signal before being transferred to the computer. The objective of an
ADC is to convert analog input data to an equivalent output digital word. Since we are
going to map a continuous value (infinite number set) to a limited number set, there will
be overlap for different input continuous values. Figure 2.42 shows the converting

50

Chapter No. Title

characteristic of a 3-bit ADC. Figure 2.42 shows that the resolution of an ADC depends
on the number of bits of the ADC. We can determine resolution using the same Eq. (2.69)
as the DAC.

111
110

Digital output

101
100

011

010
001

000
0/8

1/8

2/8

3/8

4/8

5/8

6/8

7/8

Vref

Figure 2.42 Converting characteristic of 3-bit ADC.

Many circuit configurations could be employed in an ADC, such as a counter

ADC, dual slope ADC, successive approximation ADC, or parallel ADC. Here we
present the successive approximation ADC.
Figure 2.43 is the block diagram for a successive approximation ADC. It
consists of a comparator, a DAC, and digital control logic. When the conversion begins,
it assumes the MSB (most significant bit) to be 1 and other bits to be 0. This digital value
is sent to the DAC and generates an analog value of 0.5 Vref. The comparator compares
this generated value with the input analog signal. If the input is higher than the generated
value, the MSB is indeed 1. Otherwise, the MSB is changed to 0. This process is repeated
for the remaining bits.

Section No. Title

Clock

Vin

51

Digital
control
logic

+
_

Digital
output
DAC converter

Figure 2.43 Block diagram of a typical successive approximation ADC.

Figure 2.44 shows the possible conversion paths for a 3-bit converter. The total
number of clock cycles required to convert an n-bit word is n.
Vout
111
110
101
100

1/2 Vref

011
010
001
0

111
110
101
100
011
010
001
000

2
3
4
Conversion cycle

Figure 2.44 The possible conversion paths of a 3-bit successive approximation ADC.

2.5 Digital signal processing

Digital signal processing uses discrete-time technology to process continuous-time
signals. A continuous signal is converted into a sequence of samples by the ADC. This

52

Chapter No. Title

sequence is processed by algorithms to yield an output sequence. After discrete-time

processing, the output sequence is converted back to a continuous-time signal by a DAC.
In some cases, the input samples can be stored in memory and the processing
will be implemented after all the samples are obtained. But in other digital signal
processing, real-time operation is often desirable, meaning that the system is
implemented so that samples of the output are computed at the same rate at which the
continuous signal is sampled.
Compared with analog signal processing, digital signal processing has some
advantages: (1) A digital filter is highly immune to interference because of the way it is
implemented (digital signals are far less susceptible to noise than analog signals). (2)
Accuracy is dependent on round-off error, which is directly determined by the number of
bits used to represent the digital signal. (3) With the help of modern computers, digital
signal processing is much more flexible and powerful than analog signal processing. (4)
Performance of a digital signal system is minimally affected by environmental factors
such as temperature variation, component aging and power supply fluctuations.
Because of these advantages, digital signal processing is widely used in
communication systems, radar and sonar, speech and video coding and enhancement, and
biomedical engineering.
There are some limitations of digital signal processing. The greatest drawback is
its relatively slow speed. The speed is affected by two factors. One is the sampling speed
of the ADC and DAC. The fastest ADCs and DACs work at frequencies in the range of
tens of megahertz, while analog signal processing can work at frequencies in the range of
a gigahertz. The other is the speed of the signal processing hardware. The speed of analog
signal processing is limited by the delay of the circuits. In digital signal processing, it
depends on the complexity of the algorithms and the clock speed of the processor.
2.5.1 Digital signals

Analog signals are continuous in time and amplitude. Digital signals are discrete in time
and amplitude. They are represented by a sequence of numbers, x, in which the nth
number in the sequence is denoted x[n], where n is an integer. Such sequences are usually
obtained by the periodic sampling of an analog signal xa(t). In the sampled sequence, the
value of the nth sample in the sequence is equal to the value of xa(t) (rounded to the
precision of the ADC) at time nT, or
x[n] = xa (nT )

(2.70)

where T is the sampling period and its reciprocal is the sampling frequency. As an
example, Figure 2.45(a) shows a segment of a sine wave and Figure 2.45(b) is the
sampled sequence of the sine wave.

Section No. Title

53

Amplitude

0
1

4
(a)

Time

Amplitude

0
1

10

15

20

25

Sample
numbers
30

(b)
Figure 2.45 (a) Continuous signal. (b) Sampled sequence of the signal in (a) with a sampling
period of 0.2 s.

2.5.2 Sampling theorem

To convert a continuous time signal into a discrete time signal and process it with a
microprocessor, we are assuming that we can represent a continuous time signal by its
instantaneous amplitude values taken at periodic points in time. We are also assuming
that we will be able to reconstruct the original signal perfectly with just these sampled
points.
The sampling theorem, developed by Shannon, guarantees that the original
signal can be reconstructed from its samples without any loss of information. It states that,
for a continuous band-limited signal that contains no frequency components higher than
fc, the original signal can be completely recovered without distortion if it is sampled at a
frequency of at least 2fc. A sampling frequency fs, which is twice the highest frequency in
a signal, is called the Nyquist frequency.
Suppose the original signal can be represented by function Xc(f) in the frequency
domain and its highest frequency component is fc. When it is sampled periodically at
frequency fs, then the sampled signal can be represented as (Oppenheim and Schafer,
1989)
+

X s ( f ) = fs X c ( f kfs )
k =

(2.71)

54

Chapter No. Title

where Xs(f) is the frequency representation of the sampled sequence. Figure 2.45 is a
sketch of the sampling process. We know the sampled signal consists of periodically
repeating copies of Xc(f). These copies are shifted by integer multiples of the sampling
frequency and then superimposed to produce the periodic samples. Figure 2.46 shows
that if we use a low-pass filter in which fn is the corner frequency, we can reconstruct the
original by passing the sampled signal through the low-pass filter (fn is between fc and
fs fc).
In Figure 2.46, if the frequency fs fc is less than fc, the copies of the original
signal overlap for some of the frequency components. This overlap will corrupt the
original signal and we will not be able to reconstruct the original signal from the sampled
signal. To guarantee that we can reconstruct the original signal, we must meet the
condition of the sampling theorem
fs fc > fc

or f s > 2 f c

(2.72)

Here we present an example of a moving window digital filter for an ECG signal.
In some ECG signals, there is the higher frequency electromyogram (EMG)
which interferes with the ECG signal. We can use a low-pass digital filter to attenuate the
EMG using the following equation
y[n] =

1
( x[n] + 2 x[n 1] + x[n 2])
4

(2.73)

We can also change the coefficients of the digital filter to get different results or
to use it for other applications. This is more flexible than the analog filter, which would
require a change in components or even construction of a new circuit.

Original

55

Magnitude

Section No. Title

fC

Frequency

Sampling
function

Magnitude

(a)

fS

2fS

Frequency

2fS

Frequency

Sampled
signal

Magnitude

(b)

fSfC

fC

fS

Low-pass
filter

Magnitude

(c)

fn

Frequency

Reconstructed
signal

Magnitude

(d)

fC

Frequency

(e)
Figure 2.46 (a) Spectrum of original signal. (b) Spectrum of sampling function. (c) Spectrum of
sampled signal. (d) Low-pass filter for reconstruction. (e) Reconstructed signal, which is the same
as the original signal.

2.6 Microcomputers
The computer is one of the most important inventions of the twentieth century. In the last
twenty years, computer technology, especially microcomputer technology, has had a
significant impact on bioinstrumentation.
Computers are divided, by their size and use, into several categories:
supercomputers and mainframe computers, workstations, microcomputers (or personal

56

Chapter No. Title

computers), and microprocessor-based (or micro-controller) systems. The PC and

microprocessor-based systems are generally chosen for bioinstrumentation.
2.6.1 Structure of a microcomputer

A microcomputer is composed of its CPU (central processing unit), memory, I/O devices,
and buses, as shown in Figure 2.47.
Bus (address, data, control signal)

ROM

RAM

CPU

I/O

Figure 2.47 General block diagram of a microcomputer system (arrows represent the data flow
direction).

The CPU plays an important role in a computer. It fetches, decodes and executes
instructions and controls the performance of other modules. The most commonly used
CPU in microcomputers is the Intel Pentium 4, which contains 42 million transistors on
one chip.
Read-only memory (ROM) can only be read by the CPU and random-access
memory (RAM) can be read and written to by the CPU. The core programs, such as startup instructions and interface instructions for the CPU, are generally stored in ROM.
ROM retains these programs even if there is no power. Because it holds basic programs,
a modern microcomputer contains tens of kilobytes of ROM.
The system software and applications are loaded to RAM for processing. Data
and instructions are also stored in RAM. With advances in VLSI technology, the capacity
of RAM becomes greater while the price keeps decreasing. In 2002, most computers have
more than 256 megabytes (MB) of RAM, with some having several gigabyte (GB)
memory .
The bus is a communication path used to transfer information (such as
instructions, data, addresses or control signals) between the modules that make up the
system. There may be one or more functionally separate buses linking the various
microcomputer modules.
The clock is an oscillation signal generated to synchronize the performance of
the modules in the system (which is not shown in Figure 2.47). The faster the clock speed,
the faster the computer system. The first IBM PC had an Intel 8088 with a clock speed of
4.8 MHz. In 2002, the Intel Pentium 4 has a clock speed of 2.2 GHz. Intel or other
companies will manufacture much faster chips in the future.
I/O (input/output) devices are used by the computer to input and output
information. These devices can be divided into several categories.
The most common input devices for a microcomputer are the keyboard and
mouse. A user gives the computer instructions or data through these devices. The
computer can display the result on the output devices (monitors or printers) for the user.

Section No. Title

57

Mass storage devices include floppy disk drives, hard disk drives, CD-ROM
drives and tape drives. They are used to permanently store software and data on mass
storage media.
A computer can be connected to other computers via a network. Network
devices include a modem or network card, and a network. When connected to the
network, computers can share resources (such as devices, software and databases), which
improves the performance of connected computers.
In bioinstrumentation, ADC and DAC devices are widely used to collect the
data for processing and to send out signals to control other equipment.
2.6.2 Microprocessor-based systems

The structure of microprocessor-based systems is similar to that of the microcomputer

except that it contains fewer modules. For example, a clinical ECG monitoring system
contains a microprocessor, ROM for a program, RAM for data and an I/O port for ECG
signals. It does not have mass storage (a floppy disk or hard disk), keyboard, or monitor
for display, but may have some keys to choose functions or a small printer to print out the
ECG signals. When something abnormal occurs, it gives out an alarm from a speaker and
light signals.
Compared to a microcomputer, a microprocessor-based system contains less
memory (tens of kilobytes, depending on the application), and fewer devices. The
advantages are its small size and flexibility. Some microprocessors even have ADC/DAC
circuits, communication circuits and other devices on the chip so that one single chip
satisfies all the needs of the application.

2.7 Software and programming languages

Section 2.6 discussed the components of a computer system. The computer and its
peripherals (printer, keyboard, monitor, etc.) describe the hardware of the computer
system.
Besides the hardware for bioinstrumentation, software plays a critical role to the
performance of the system. The computer itself can do nothing without software. The
types of software are limited by the hardware system. Figure 2.48 shows that there are
three levels of software between the hardware and the real world: the operating system
(UNIX, DOS, Windows, and Macintosh), the support software and the application
software. It is the application software that gives the computer its bioinstrumentation
abilities.

58

Chapter No. Title

Problem
Application software
Editor

Complier

Linker

Operating System
CPU

Figure 2.48 Three levels of software separate the hardware of microcomputer from the real
problem.

Two aspects should be considered for developing a biomedical system: (1) the
choice of operating system to support the task, and (2) the choice of the programming
language to implement the application. There are many combinations of operating
systems and programming languages that are able to address the problem, but some
choices are better than others for certain applications.
2.7.1 Operating system

The operating system (OS) is a set of procedures that handle the resources of the
computer system. The resources include the CPU, file system, display, printer and so on.
The OS manages the file system control, CPU control, memory control and system
security. It provides a virtual machine so that the user just deals with the operating system
instead of the computer hardware. The operating system also provides some frequentlyused functions so that users can work more efficiently.
An example of the type of task an OS handles is file system management. In a
computer, there are several storage media such as a floppy disk, hard disk, or tape. When
saving files, we have to consider the following: (1) the mechanisms of these devices are
different, so we should use different procedures to write our files to each one, (2) for a
device, we have to decide the format of the file and location of the file, (3) we have to
determine a method for file indexing so that we can obtain access to the file. The OS
provides an interface between the user and media. A user just asks the operating system
to save the file. The OS takes care of the details automatically. Without the OS, computer
users have to communicate with the computer hardware directly by writing some code,
which would take a lot of effort.
Below is a brief discussion of three common operating systems: Unix,
DOS/Windows, and Mac.
Unix is widely used on mainframe computers and workstations, although some
versions of Unix are available on microcomputers. It is very flexible, provides the
capability to maximally manipulate a computer system, and gives excellent control of
input/output and other facilities. It also provides for multiple simultaneous users to do
multiple simultaneous tasks (i.e. it is a multiuser, multitasking operating system).
Because of this great flexibility, Unix requires considerable expertise to use all of its

Section No. Title

59

capabilities. Different hardware companies (such as HP, IBM and others) provide their
own versions of Unix for their products. However, all versions of Unix are very similar.
The Macintosh is a computer hardware/software system by Apple. It is designed
for ease of use and for graphics-oriented applications. Macintosh is equipped with a very
easy to use graphical user interface (GUI) by sacrificing a great deal of direct user control
of the hardware system. The concept was to produce a personal computer that would be
optimal for running applications instead of writing new ones. Without training, an
individual can sit down and quickly learn to use the Macintosh because of its intuitive OS.
MS-DOS was designed by Microsoft for the IBM PC and compatibles. It
became the most popular operating system on personal computers because of the wide
use of the IBM PC. In the 90s, MS-DOS evolved to Windows system (3.1, NT, 2000,
XP), which has graphical user interface. In DOS/Windows, a user can still control the
hardware of the system, but it is not as flexible as the Unix system.
A typical need for bioinstrumentation is the ability for real-time processing. In
other words, we need the results when we collect the data. Sometimes this is critical,
especially in the intensive care unit of a hospital. In most cases, biomedical applications
require a single-user system that gives maximal performance from the computing power,
and the DOS/Windows system satisfies this requirement. The Unix system is not
desirable because of its multi-user/multitask property. The Macintosh system is not
desirable, as the user cannot control the hardware structure.
2.7.2 Programming languages

The OS takes care of talking to the hardware but the user still has to give the computer
instructions on how to implement the solution (i.e. program the computer).
The instructions for the computer are in binary format and are machine
dependent, which means that binary codes can be used on only one kind of computer. For
example, the binary code for PC machines that use DOS cannot be used on Macintosh
machines.
Since instructions for the CPU are binary, the final codes for the computer
should be in binary format. When the computer was first invented, engineers had to
remember instructions in binary code and write their own programs in binary format.
After 40 years of development, there are now many programming languages, some of
which are similar to our native languages. The codes developed in the programming
language are called source codes. A complier is used to translate the source codes to
binary codes that computers can understand. One advantage of programming languages is
ease of use. Another advantage is transportability, which means user can use the source
codes on different machines with different operating system provided that the language
compiler is available in the different operating systems.
There are numerous programming languages, but all languages can be classified
in one of two categories.
One category is low-level languages, which includes assembly language. Every
instruction in assembly language corresponds to a binary instruction code for the CPU.
Therefore, it is known as a low-level language. Assembly language can achieve the
greatest run-time performance because it provides the direct manipulation of the

60

Chapter No. Title

processor. However it is also the most difficult language to write since we have to think
about the problem in the way the processor does. Another drawback is that the program is
machine dependent. Once we want to use another system, we have to write the code
again from scratch.
The other category of programming languages is high-level languages. The
instructions are similar to the way we think of the problem and programs are relatively
easy to write. High-level languages make developing programs much easier. The tradeoff
is that the programs usually are not as efficient as those in assembly language, especially
for real-time processing.
The most widely used high-level languages are FORTRAN, Pascal and C/C++.
FORTRAN was developed for scientific calculation (FORmula TRANslation). Pascal
was designed for teaching students structural programming techniques.
The C language is the most popular language and is used to develop Unix
systems. It has the characteristics of a high-level language, but also contains functions
and interfaces for low-level programming. C++ is a modern version of the C language
and allows object-oriented structure. C/C++ is the current language of choice for realtime programming, with some parts of the code possibly written in assembly language for
increased efficiency.
There are two other special languages worth mentioning. LabVIEW is a visual
computing language available for many operating systems. It is optimized for laboratory
applications. Programming is accomplished by interconnecting functional blocks that
represent processes such as data acquisition and analysis or virtual instruments. Thus,
unlike traditional programming by typing command statements, LabVIEW is a purely
graphical, block diagram language. In most cases, LabVIEW is used to acquire and
process data in the laboratory and in some industrial applications.
MATLAB is general purpose software for matrix manipulation. It is used for
matrix calculation and also has many packages containing various functions such as
signal processing, image processing, data analysis and so on. MATLAB is very useful for
digital signal processing, which in turn is very important for biomedical instrumentation.
2.7.3 Algorithms

An algorithm is a list of steps that solve a given problem. Such a description should be
precise enough to be used in a completely automatic manner (i.e. by a computer). It is the
programmers responsibility to ensure the correctness of the algorithm.
There are several aspects of the correctness of an algorithm. For every satisfying
input, the results of the algorithm should be correct. Also for every satisfying input, the
algorithm should execute to the end and not abort in the middle. If these conditions are
not met, the system will produce incorrect results, especially for real-time processing.
The algorithm should also terminate in a finite number of steps. An algorithm that needs
an infinite number of steps is useless.
Another consideration for an algorithm is its efficiency. Execution of an
algorithm on a computer requires a certain amount of computer resources as well as a
certain amount of computing time. Before implementing an algorithm on a computer, we
have to investigate whether it is feasible to use the algorithm to solve the problem. A lot

Section No. Title

61

of known algorithms grow too computationally expensive as the size of the input
increases. For many real-time processing applications in bioinstrumentation, execution
time and memory requirement are important considerations in system design.
2.7.4 Database systems

A database is a collection of stored operational data used by applications. Database

systems are composed of three parts. The database itselfa collection of data stored on
some storage media; A set of ordinary programs for users to get access to the data for
performing the functions such as retrieving, updating, inserting and deleting; The
database is integrated, which means that it contains data for many users. Different users
are concerned with only a small portion of the data, but many different users may share
pieces of data.
Why are databases important? One general answer is that they provide users
with centralized control of the operational data. If every user in a group or each
application has its own private file, then the operational data would be widely dispersed.
A typical use of database systems is in hospitals. All the information about a patient is
stored in computers from which doctors can retrieve the data. Even if it is the first time
that patient is admitted to a certain hospital, doctors can obtain his or her former
information from the other hospitals using a database system.
There are several advantages of having centralized control of data.
First, the amount of data redundancy in the stored data can be reduced. If each
application has its own private files, this often leads to considerable redundancy in the
stored data. For example, a patients files for different doctors all contain his or her name,
social security number, address, age, gender and so on. With central control, this data can
be stored only once and shared by different doctors.
Second, the data can be shared. This means any doctor can use the database if he
or she is granted access.
Third, data security restrictions can be applied. The database administrator can
assign access to users for certain data. This can protect data from abuse.
Another important advantage of centralization is the data independence of
databases. Most applications are data-dependent, which means the way in which the data
are stored and the way in which they are accessed are both dictated by the application.
With data independence, the application does not need to know the storage
structure of the data and access strategy. By using the same commands, the application
can get access to different databases without the knowledge of the physical storage and
access format. Different applications can have different views of the same data and the
data administrator responds to different requirements automatically.
The architecture of the database determines the performance of the application.
The data model is the heart of any database system. There are different ways to
implement a database. The three best-known approaches are the relational approach, the
hierarchical approach and the network approach.

62

Chapter No. Title

2.8 Display devices

For improved understanding of the processes of the instrument or of an object, it is
important to display the results as either characters or graphically.
The most common device for displaying electric signals is the oscilloscope. The
main part of an oscilloscope is a cathode ray tube (CRT). Figure 2.49 is a sketch of a
CRT.
The cathode emits electrons. Electrons are accelerated by the axial high voltage,
hit the screen, and light is emitted due to fluorescence. The electrons pass through a
transverse electric field before hitting the screen and deflect at a certain angle, which is
controlled by an electric signal. The deflection control unit controls the location of the
spot on the screen. For an oscilloscope, the horizontal deflection control signal is a time
swept signal and the vertical deflection control signal is the signal we display.
The heart of a TV monitor is also a CRT, but it is a little different from the CRT
for the oscilloscope. In a TV monitor, the signal received from the antenna controls the
number of electrons emitted from the cathode and the brightness of the spot on the screen.
If many electrons hit the spot, it will emit more light and we observe that the location is
white. If no electrons hit the spot, it emits no light and we observe it is black. The
deflection control unit controls the location of the spot (scanning) and we can see an
image on the screen. For a color TV monitor, there are three colors (red, blue, green) of
fluorescent material on the screen. These three colors compose the different colors we
observe.

Cathode

Deflection
control

Screen

Figure 2.49 Sketch for cathode ray tube (CRT). There are two pairs of electrodes to control the
deflection of the electron, but only one pair is shown.

Section No. Title

63

2.9 Recording devices

In many applications, we want to save the data permanently as record. A strip chart
recorder is a common electromechanical device that records an electric signal versus time
in the old days. A strip of paper travels at a constant speed (i.e., 5 cm/s, 10 cm/s) and an
ink pen is in contact with the paper. The pen arm moves freely normal to the direction of
paper movement and its movement is controlled by the input electric signal. It records
(plots) the electric signal on the paper. Chart recorder is suitable for low frequency
signals.
A tape recorder is also a common device for recording analog signals. The
device is similar to the tape recorder we use in everyday life. The main difference being
that we record the input electric signal instead of music. We can retrieve the signal from
the tape and send it to a device to display the signal or make a measurement.
In recent years, technological advances have made digital techniques more
widely used in recording signals.
The digital oscilloscope can sample the signal at a high sampling speed and
display the signal. It also has an interface that can be used to save data on a floppy disk or
to send data to a computer, which can save the file. The computer printer output replaces
the strip chart recorder in many biomedical instruments such as ECG, pulse oximeter and
spirometry.
Another commonly used device is a PC data acquisition system, such as
LabVIEW Biobench. A personal computer uses an ADC unit to sample the signal and
save the signal as a file. It can also display the signal on the screen for manual or
automatic measurement, in which case it is similar to the oscilloscope. The advantage of
this kind of system is the computing power of the computer can be employed to process
the data. This makes the system more flexible than an oscilloscope.

2.10 References
Alexandridis, N. A. 1984. Microprocessor System Design Concepts. Rockville, MD:
Computer Science Press.
Banachowski, L., Kreczmar, A. and Rytter, W. 1991. Analysis of Algorithms and Data
Structure. Reading, MA: Addison-Wesley.
Bobrow, L. S. 1987. Elementary Linear Circuit Analysis. New York: Holt, Rinehart and
Winston.
Date, C. J. 1977. An Introduction to Database Systems. Reading MA: Addison-Wesley.
Hansen, P. B. 1973. Operating System Principles. Englewood Cliffs, NJ: Prentice-Hall.
Hilburn, J. L. and Julich, P. M. 1979. Microcomputers/Microprocessors: Hardware,
Software and Applications. Englewood Cliffs, NJ: Prentice-Hall.
Lurch, E. N. 1963. Electric Circuits. New York: John Wiley & Sons.
Oppenheim, A. V. and Schafer, R. W. 1989. Discrete-time Signal Processing. Englewood
Cliffs, NJ: Prentice-Hall.
Tompkins, W. J. (ed.). 1993. Biomedical Digital Signal Processing. Englewood Cliffs,
NJ: Prentice-Hall.

64

Chapter No. Title

Webster, J. G. (ed.) 1998. Medical Instrumentation: Application and Design. 3rd ed. New
York: John Wiley & Sons.

2.11 Problems
2.1
2.2
2.3
2.4
2.5
2.6
2.7
2.8
2.9
2.10
2.11
2.12
2.13
2.14
2.15
2.16
2.17
2.18

One ampere flows through an electrosurgical dispersive electrode that has a

resistance of 50 . Calculate the power dissipated. Discuss the possibility of burns
under the electrode.
Calculate the equivalent resistance for a 400 and 1 k resistor in series.
Calculate the equivalent resistance for a 400 and 1 k resistor in parallel.
Design a resistive attenuator to convert a 20 V voltage from a signal generator to a
10 mV voltage for an amplifier.
An operational amplifier has limited current output of 20 mA. Calculate the time to
charge a 1 F capacitor from 0 to 10 V.
From a 2 kV source in series with a 20 k resistor, calculate the time required to
charge a 100 F defibrillator capacitor to 1.9 kV.
Calculate the time response of a 10 k, 30 mH high-pass and low-pass RL circuits
(similar to the two cases of RC circuit, but = L/R) to a 1 V step input.
Design an inverting amplifier with a gain of 10. If a voltage of 2sin(2t) is
applied, sketch the waveform of the output and label axes.
Design a noninverting amplifier with a gain of +10. If a voltage of 2sin(2t) is
applied, sketch the waveform of the output and label axes.
In Figure 2.30, R1 = 1 k, the two R2 are 10 k and 9 k respectively (the one
connected to ground is 9 k). Calculate the CMRR.
Design a comparator to detect an amplified electrocardiogram whose Q wave is 0
V and R wave is 4 V.
A low-pass filter with R = 10 k and C = 300 pF is placed before an ECG
amplifier. Calculate by what factor it reduces 1 MHz interference from an
electrosurgical unit.
Calculate the number of bits to obtain a resolution of 1/1000 Vref for an ADC.
Explain the advantages and disadvantages of digital signals versus analog signals.
Define Nyquist frequency. For a signal that contains frequency components up to 1
kHz, calculate the minimum sampling frequency.
Several students are attempting to digitally process a sinusoidal signal that repeats
every 250 s. The students randomly select a DSP that takes 360,000 samples per
minute. Will they be successful in processing the signal correctly? Why or why not?
Calculate the minimum bit size and sampling speed for a DSP to handle an
incoming signal with a range from 0 to 5 V, frequency components between 25 and
250 Hz and sensitive to 1 mV.
Explain the function of the following devices in a microcomputer: CPU, RAM,
ROM, I/O port, data bus.

Section No. Title

65

2.19 Give two examples for each category of software: application software package,
high-level programming language, low-level programming language. Also state on
which platform each operates.
2.20 Describe how a CRT functions.
2.21 Go to www.bmenet.org/BMEnet/. Click on jobs. Find an industrial (not university)
job and describe in a few sentences.
2.22 A radio-frequency (RF) catheter is introduced into the heart to inject power at 500
kHz to ablate (destroy) tissue that causes tachycardia. This causes interference in
the electrocardiogram (ECG). Design a filter that reduces this interference to
0.0005 of its former level.

3
Analysis of Molecules in Clinical Medicine
Mat Klein

Methods used to measure molecules in clinical chemistry, toxicology, and pulmonary

medicine, have a tremendous impact on the fields of bioengineering and medicine. The
ability to measure certain molecules in the body enables us to better understand the biological and chemical processes occurring in the body. This chapter presents various techniques used to measure different molecules in the body, along with the importance and
impact of their measurement.

3.1 Spectrophotometry
Photometry is based on measurements of electromagnetic energy emitted, absorbed, or
transmitted under controlled conditions. Spectrophotometry and flame photometry are
two different types of photometry commonly used to determine the concentration of light
absorbing molecules in a solution. Spectrophotometry is based on the use of light intensity measurements at a particular wavelength or a discrete portion of wavelengths to determine the concentration of the desired molecules in solution by the amount of radiant
energy absorbed by the desired molecules.
3.1.1 Components
The principles of spectrophotometry can be understood by examining the single beam
spectrophotometer. Figure 3.1 shows that the components of the single beam spectrophotometer include a light source, monochromator, cuvette, detector, and readout device.
Radiant energy sources
The purpose of the light source in a spectrophotometer is to provide incident light of sufficient intensity to the sample. The wavelength and intensity of incident light desired determines what light source to use. The most common source for substances that absorb

3-1

3-2

Chapter 3 Analysis of Molecules in Clinical Medicine

radiation in the visible, near infrared, and near ultraviolet regions, is a glass enclosed
tungsten filament. These distinctions between the spectrum are made since silica, used to
make cuvettes, transmits light effectively at wavelengths greater than 220 nm. To increase the lifetime of the tungsten filament, there is usually low-pressure iodine or bromine vapor in the bulb. The tungsten light bulb does not supply enough radiant energy for
measurements below 320 nm. High-pressure hydrogen or deuterium discharge lamps are
sufficient for measurements in the near ultraviolet region. At wavelengths below this, the
emission is no longer continuous. Two advantages deuterium lamps have is that they produce about three times the light intensity of hydrogen lamps and have a longer lifetime.
Xenon arc or high-pressure mercury vapor lamps provide high levels of continuous ultraviolet illumination. These lamps may require thermal insulation with or without auxiliary
cooling to protect surrounding components since they become extremely hot during operation.
Exit slit
Entrance slit

Detector
Red

Prism

I0

Readout
device

Violet
Cuvette

Light source

Monochromator

Figure 3.1 Block diagram of a single beam spectrophotometer. The prism serves as the dispersing
device while the monochromator refers to the dispersing device (prism), entrance slit, and exit slit.
The exit slit is moveable in the vertical direction so that those portions of the power spectrum produced by the power source (light source) that are to be used can be selected.

Monochromator
A monochromator is a system for isolating radiant energy of a desired wavelength. The
term monochromator refers to the dispersing device and associated slits and components
used to isolate the desired wavelength. A monochromator commonly used in spectrophotometers uses prisms or diffraction gratings. Both components separate white light into a
spectrum from which the desired wavelength may be chosen. A prism separates white
light into a continuous spectrum by refraction. As white light passes through the prism,
shorter wavelengths are refracted, or bent, more than longer wavelengths. Although
longer wavelengths are closer together than shorter wavelengths, since refraction is
nonlinear, with the proper components, a narrow band or the desired spectrum can be
isolated. A diffraction grating separates white light, such as that produced by a tungsten
filament, into a continuous linear spectrum. Diffraction gratings used in spectrophotometers consist of many closely spaced parallel lines on a substrate such as glass covered

3.1 Spectrophotometry

3-3

with a polished aluminum or aluminumcopper alloy. When radiation strikes the grating,
light rays bend around the sharp edges of the closely spaced parallel lines. The amount of
bending is dependent upon the wavelength of the light. A tiny spectra is produced for
each line of the grating. Wave fronts are formed as the light waves move past the corners.
Wave fronts that are in phase when they cross reinforce each other, while wave fronts
that are out of phase cancel out, leaving a complete spectrum from which to choose a
narrow band of wavelengths or particular wavelength.
Cuvette
Spectrophotometry ascertains the absorption of the desired molecules (the solute) in a
solvent. A cuvette, also referred to as an absorption cell, holds the solute and solvent.
Cuvettes can be round, square, or rectangular, and have a light path of constant length,
most commonly 1 cm. For measurements made in the visual range (above 340 nm), cylindrical test tubes are sufficiently accurate. They are made from glass tubing, are not
perfectly round or polished, and contain surface aberrations. For measurements below
340 nm, square or rectangular cuvettes made of quartz or silica, which are free of optical
aberrations, are required.
Detector and readout device
There are two requirements for a detector of radiant energy in a spectrophotometer. The
photosensitive detector must have a linear response and be sensitive enough in the part of
the spectrum it is being used in. The most common devices used to detect the amount of
radiant energy leaving the cuvette include barrier layer cells, photodiode arrays, and photomultiplier tubes. These devices convert electromagnetic energy to electric energy,
which can then be measured.
A readout device displays the electric energy from the detector onto some type
of scale such as absorbance or transmittance. The two types of readout devices are the
direct reading system and the null point system. In a direct readout meter, there is a linear
relationship between milliamperes and %T (percent transmittance) and a log relationship
between millivolts and absorbance. This characteristic makes direct readout systems fast
and simple. In a null point system, the meter is calibrated (zeroed) by a potentiometer. In
a null point system, the absorbance, transmittance, or any other arbitrary scale is fitted to
the potentiometer scale.
3.1.2 Theory
The determination of the concentration of a light-absorbing substance in a solution using
a spectrophotometer is based on the discoveries of four individuals. Bouguer and Lambert noted that transmittance of light through an absorbing material, such as that contained in a cuvette, decreases exponentially with an increase in the light path through the
cuvette. Beer and Bernard observed that the concentration of a substance in solution [less

3-4

Chapter 3 Analysis of Molecules in Clinical Medicine

than 102 M (molar)] is directly related to its absorbance. An amalgamation of these two
discoveries is known as Beers Law, as stated by Bouguer: Equal thickness of an absorbing material will absorb a constant fraction of the energy incident upon it. (Wheeler,
1998) This relationship is

I = I 010 aLc

(3.1)

where
I0= radiant power arriving at the cuvette
I = radiant power leaving the cuvette
a = absorptivity of the sample (extinction coefficient)
L = length of the path through the sample
c = concentration of the absorbing substance

Transmittance is the ratio of light intensity leaving the cuvette I to light intensity entering
the cuvette I0

T=

I
I0

(3.2)

Absorbance (A) is exponentially related to the reciprocal of transmittance.

I
100
A = log 0 = log
= 2 log(%T )
I
%T

(3.3)

As the concentration of the substance in solution increases (or decreases), the transmittance varies logarithmically and inversely. The fraction of incident light absorbed by a
substance in solution at a particular wavelength is a constant characteristic of the substance called absorptivity, a. Let the length of the light path (usually the outer diameter of
the cuvette) be a constant, L, and the concentration of the substance in solution be c. The
relation of these parameters to the total absorbance is expressed as:
A = aLc

(3.4)

where a is in liters per gram times centimeters, L is in cm, and c is in grams per liter. Absorptivity is often replaced by molar absorptivity, , expressed in liters per mole times
centimeters. Epsilon is a constant corresponding to a 1 molar solution of the absorbing
substance, with a light path of L = 1 cm and a given wavelength. Hence, absorbance can
be expressed as
A = Lc

(3.5)

3.1 Spectrophotometry

3-5

If the concentration and absorbance of a substance is known at a specific wavelength and

this concentration is changed, we can calculate the unknown concentration of the substance using the absorbance of the unknown concentration and the equation:
cu Au
=
cs
As

(3.6)

where Au is the absorbance of the unknown concentration and cu is the unknown concentration.
3.1.3 Calibration

The purpose of calibrating a spectrophotometer is to make sure that the wavelength of

radiant energy emitted from the exit slit of the monochromator is within certain tolerances of the wavelength of radiant energy that appears on the wavelength selector of the
instrument. This ensures that measurements made with the spectrophotometer are accurate and reproducible.
3.1.4 Clinical laboratory tests

Table 3.1 lists common molecules measured in the clinical laboratory and their normal
and toxic levels
Molecule

Type of test

Normal levels
US units
(mg/dL)

Normal levels
SI units
(mmol/L)

Toxic levels
(mg/dL)

Total bilirubin
Indirect
Direct

Blood (serum)

0.21.0
0.20.8
0.10.3

High levels result in jaundice.

Lactate
Creatinine
Female
Male

Blood (serum)
Blood (serum)

512

3.417.1 103
3.412 103
1.75.1 103
0.51.3

Urea

Blood (serum)

1020

5397 103
62115 103
3.67.1

Glucose

Blood

70105

3.95.8

Adult male:
<50 or >400
Adult female:
<40 or >400

Sodium

Blood (serum)

136145 mEq/L

136145

<120 or >160 mEq/L

Potassium

Blood (serum)

3.55.1

<3 or >7.5 mmol/L

Lithium

Blood (serum)

NA

>2.0 mEq/L

0.61.1
0.71.2

0.81.2 mEq/L

>45
>4

>100

Table 3.1 Normal and toxic levels of various molecules in the body given in both US and European
(SI) units (Pagana and Pagana, 1995). NA denotes not available.

3-6

Chapter 3 Analysis of Molecules in Clinical Medicine

3.1.5 Microdialysis

Microdialysis is a technique for sampling the chemistry of the individual tissues and organs of the body, and is applicable to both animal and human studies (de la Pea, 2000).
The basic principle is to mimic the function of a capillary blood vessel by perfusing a thin
dialysis tube implanted into the tissue with a physiological liquid. The perfusate is analysed chemically and reflects the composition of the extracellular fluid with time due to
the diffusion of substances back and forth over the membrane. Microdialysis is thus a
technique whereby substances may be both recovered from and supplied to a tissue. The
most important features of microdialysis are as follows: it samples the extracellular fluid,
which is the origin of all blood chemistry; it samples continuously for hours or days
without withdrawing blood; and it purifies the sample and simplifies chemical analysis by
excluding large molecules from the perfusate. However, the latter feature renders the
technique unsuitable for sampling large molecules such as proteins. The technique has
been extensively used in the neurosciences to monitor neurotransmitter release, and is
now finding application in monitoring of the chemistry of peripheral tissues in both animal and human studies.

3.2 Oxygen saturation

Hemoglobin is a protein and the main component of red blood cells. Hemoglobin transports oxygen from the lungs, where oxygen tension (partial pressure of oxygen) PO2 is
high, to the tissues, where oxygen tension is low (See Chapter 6). Oxygen saturation, SO2,
is defined as the ratio of the amount of bound oxygen to the total oxygen capacity.

HbO 2SAT=

[HbO 2 ]
[RHb] + [HbO 2 ]

(3.7)

where [HbO2] is the concentration of oxyhemoglobin, and [RHb] is the concentration of

deoxyhemoglobin (hemoglobin that is not carrying O2). Oxygen saturation is calculated
as a percent or fraction.
HbO 2SAT(%)= HbO 2SAT(fraction) 100

(3.8)

Oxygen saturation measurements are made in order to determine the response to

therapeutic intervention (e.g. supplemental oxygen administration or mechanical ventilation) and/or diagnostic evaluation (e.g. the effect of exercise on O2 levels). Oxygen saturation measurements are also needed to monitor the severity and progression of some
diseases. SO2 can be determined by using spectrophotometry on a venous whole blood
sample. The extinction coefficients of both hemoglobin species are equal at the isosbestic
wavelength 805 nm, while reduced hemoglobin [RHb] is more transparent (smaller extinction coefficient) than oxyhemoglobin [HbO2] in the infrared region. An oximeter uses
a spectrophotometer to measure light transmission at two wavelengths to calculate SO2.

3.2 Oxygen saturation

3-7

By using four wavelengths, the CO-oximeter also measures carboxyhemoglobin ([COHb])

and methemoglobin ([MetHb]).
To measure SO2 noninvasively, we shine light through the finger or earlobe. But
then we measure a mixed sample of venous and arterial blood. To measure the arterial
blood alone, we use the pulse oximeter (Webster, 1997). Light emitting diodes (LEDs)
transmit light at 660 nm and 940 nm and a photodiode measures the transmission at the
minimum blood pressure (diastole) and at the maximum blood pressure (systole). By subtracting the measurements, we calculate the SO2 for only the additional blood added to
the expanded artery and this is all arterial blood.
The OxiFirst Fetal Oxygen Saturation Monitoring System is inserted into the
mothers uterus and placed against the fetuss temple or cheek when fetal heart rate indicates that the baby may be in distress. This allows labor to progress when SO2 indicates it
is safe to do so and thus minimizes the cesarean section rate (Mallinckrodt, 2001).

3.3 Bilirubin
Red blood cells are replaced approximately every 100 days, which means that every day
one percent of the bodys red blood cells, which are produced in bone marrow, are replaced. Bilirubin is waste resulting from the removal of old red blood cells.
Hemoglobin consists of four subunits. Each subunit has one chain of protein,
also referred to as globin, and one molecule of heme. Heme is made up of a single iron
atom and porphyrin, a ring shaped molecule to which is attached the iron atom. When a
red blood cell is destroyed, the body recycles the iron. The ring shaped molecule is toxic
and consequently is broken down into bilirubin. Unconjugated bilirubin is produced in
the spleen when the porphyrin is broken down. The unconjugated bilirubin enters the
blood stream and travels to the liver where it is converted to conjugated bilirubin and
subsequently excreted.
Unconjugated bilirubin is produced when red blood cells are destroyed. Abnormally high levels of conjugated bilirubin in the bloodstream result from liver disease and
can turn the whites of a persons eyes yellow as well as their skin. This condition is
known as jaundice. Neonatal jaundice, a common problem that occurs after birth, is a
result of the mothers antibodies attacking the babys red blood cells. In general, blood
samples can be taken and used to measure bilirubin concentration when diagnosing liver
and/or biliary disease.
The most common way to quantitatively measure bilirubin is based on the diazo
reaction and spectrophotometry. This technique is used to measure total and conjugated
bilirubin in blood serum or plasma. A detailed discussion of this method, also known as
the Jendrassik and Grof Technique, can be found in (Burtis and Ashwood, 1994).

3.4 Lactate
Lactate, also known as blood lactate in medicine, is the anionic form of lactic acid present in the blood. Lactic acid is a metabolic intermediate involved in many biochemical
processes including glycolysis and gluconeogenesis (the formation of new glucose from

3-8

Chapter 3 Analysis of Molecules in Clinical Medicine

noncarbohydrate sources). The measurement of lactate is important because any disease

that causes decreased tissue oxygen consumption increases the amount of lactate in the
blood and can be discovered this way. Some of these diseases include diabetes mellitus,
neoplasia (tumor growth), and liver disease. Lactate measurement is also important in
industry for the regulation and control of food products such as milk (Canh, 1993). Lactic
acidosis (very high levels of lactate in the blood) is caused by hypoperfusion: a decreased
blood flow through an organ. If prolonged it may result in permanent cellular dysfunction
and death. Lactic acid is measured from a sample of whole blood or plasma.
Lactate is most commonly measured by attaching the enzyme lactate oxidase to
a PO2 electrode (section 3.8.1) to detect oxygen consumption, or on a platinum electrode
to detect the amount of hydrogen peroxide formed. This latter reaction is
lactate oxidase

L-lactate + O2 pyruvate + H2O2

(3.9)

The enzyme lactate dehydrogenase (LDH) recycles pyruvate into lactate

LDH

Pyruvate + NADH + H+ L-lactate + NAD+

(3.10)

Coupling these two enzymes enables measurements of lactate as low as 80 nM (Canh,

1993). This is the method of choice for measuring lactate since it is simple and has high
specificity (see Chapter 1).

3.5 Creatinine
Phosphocreatine is found in muscle cells and can be used to convert adenosine diphosphate (ADP) back to adenosine triphosphate (ATP), thereby replenishing the muscle
cells energy. During this conversion, creatine is produced. Creatine is usually converted
back to phosphocreatine and the cycle starts over. However, when creatine needs to be
excreted, it is dehydrated and converted to creatinine. Creatinine has the property of not
being reabsorbed when going through the kidneys during urine production. Therefore, it
is an ideal candidate for being used to measure the condition of the kidneys.
Creatinine can be measured in the blood (serum) and in urine. Elevated levels of
creatinine in the blood result from muscle damage or as a result of strenuous physical
activity. In a person with kidney disease, creatinine builds up in the blood since it is being
produced faster than it is being eliminated. Therefore, measurement of creatinine in blood
(serum) is a rough estimate of the health of the kidneys. Measurement of creatinine from
the urine is a much better measure of kidney function (see Chapter 9) and is the most
prevalent clinical test for approximating glomerular filtration rate (the rate of filtration by
the kidneys).
The most common method used to measure creatinine is based on the Jaffe reaction
alkaline

Creatinine + Picric Acid Picratecreatinine Complex

(3.11)

3.5 Creatinine

3-9

In an alkaline (basic) medium, creatinine and picric acid form a redorange compound
whose structure has been postulated but not confirmed. The hypothesized picrate
creatinine complex forms from a creatinine to picric acid ratio of 1:1. The complex can
be measured spectrophotometrically at wavelengths between 505 to 520 nm. The concentration of the hydroxyl ion (alkaline medium) affects the rate of the complex formation.
Most methods use a 0.5 M concentration of sodium hydroxide and picric acid in excess of
stochiometric amounts so that picric acid is not the limiting reagent. One of the main
problems with the Jaffe reaction is that it is nonspecific when used to measure creatinine
in plasma. Molecules that interfere with the specificity of the Jaffe reaction include glucose, protein, ascorbic acid, acetone, and pyruvate. As a result, several modifications
exist which increase the specificity of the reaction (Burtis and Ashwood, 1994). However,
many interference problems are still unresolved.

3.6 Urea
Urea, NH2CONH2, a nitrogen containing molecule, is a metabolic product of breaking
down proteins. Figure 3.2 shows how urea is formed in the liver.
Protein

Amino acids

Proteolysis

Liver

NH3
Ammonia

Removal of NH2

Urea

Enzymatic
synthesis

Figure 3.2 Origin of urea in the body (Burtis and Ashwood, 1994). The liver produces urea by first
breaking proteins down into their building blocks, amino acids, by a process that breaks the peptide
bonds between the amino acids (proteolysis). The amino group (NH2) of amino acids is removed
and ultimately used to form ammonia (NH3) and urea.

Over 90% of urea is excreted through the kidneys to the urine. The body produces urea to
rid itself of excess nitrogen.
Ammonia is produced by the liver as a waste product of gluconeogenesis and
the liver converts it into urea. Urea is then transported in the blood to the kidneys as
blood urea nitrogen (BUN). Although the urea nitrogen measurement is often referred to
as BUN, it is never measured from whole blood. Urea nitrogen is most often measured
from blood serum (watery fluid separated from coagulated blood) and sometimes plasma.
An above normal amount of urea in the blood is an indicator of decreased kidney function, and therefore possibly kidney disease.
The two primary methods of measuring urea nitrogen are spectrophotometric.
The first method measures urea indirectly by quantifying the concentration of the ammonium ion spectrophotometrically. Urea is hydrolyzed by water in the presence of urease
(an enzyme) and the resulting ammonium ion is quantified
urease

NH 2 CO NH 2 + H 2 O 2NH 2 + CO 2
2 NH 4 + + CO3 2 (3.12)

3-10 Chapter 3 Analysis of Molecules in Clinical Medicine

Two ways of quantifying the ammonium ion concentration are Berthoelots reaction and
the enzymatic assay with glutamate dehydrogenase. An electrochemical approach exists
but is too slow to be used repeatedly (Burtis and Ashwood, 1994).
The second method measures urea directly. It is based on the Fearon reaction,
and can measure urea in urine and plasma. It has been used in several automated systems,
such as the Dupont Dimension 380. The reaction is
H + , Heat

Urea + Diacetyl Diazine + H 2 O

(3.13)

Diazine can be measured using spectrophotometry and absorbs electromagnetic energy at

540 nm. Since diacetyl monoxime is unstable in the presence of water, diacetyl can be
generated by the condensation reaction
H+

DiacetylMonoxime + H 2 O
Diacetyl + Hydroxylamine

(3.14)

3.7 Glucose
Glucose is the main source of energy for all organisms. Diabetes mellitus is a group of
metabolic disorders of carbohydrate metabolism in which glucose is underutilized, producing hyperglycemia (high blood sugar levels). The two types of diabetes mellitus are
insulin dependent diabetes mellitus (type I) and non-insulin dependent diabetes mellitus
(type II). When a person has high blood glucose, glucose shows up in the urine. Normally,
sugar is not present in the urine.
Insulin is one of the hormones that controls whether or not glucose is going to
be taken out of storage and put into the blood, or vice versa. In type I diabetes, the body
does not produce insulin because there is some destruction of the pancreatic islets that
produce it. Type II patients produce insulin, but cells do not recognize it (i.e. they have
defective insulin receptors). Type I is a more severe form of diabetes mellitus, is generally seen in children, and has to be treated with insulin. Type II is seen in much older
people, and for many, careful control of diet and exercise will be enough for treatment. In
more severe cases, insulin is taken.
Self-monitoring of blood glucose is required for diabetic patients, especially
insulin dependent, in order to maintain normal blood glucose levels (glycemia). When an
abnormal amount of glucose is present in the blood, the individual needs to correct the
abnormality to avoid short and long term health complications. By regulating blood glucose levels, patients are mimicking the body by providing themselves with the correct
amount of insulin. Insulin dependent patients need to measure their blood glucose levels
about three times a day.
One reason the level of blood glucose needs to be tightly regulated is that glucose is the only source of energy neurons can consume (they do not have the enzymes to
consume anything else). Low blood glucose levels result in hypoglycemia. When this
occurs, an individuals neurons have no source of energy and if low enough, the person
will go into a coma and die. High blood glucose levels result in hyperglycemia. When
blood glucose becomes too high, the glucose molecules will denature (alter the shape of)
proteins, such as collagen and hemoglobin, throughout the body. Collagen attaches to the

3.7 Glucose

3-11

lining of blood vessels (the basement membrane). When glucose denatures collagen,
blood vessels are destroyed. This leads to decreased blood flow in the arms and legs
(lower perfusion). When glucose denatures proteins associated with neurons, nerve damage occurs and results in a persons inability to feel. Diabetes mellitus is the leading
cause of amputation because decreased blood flow causes tissue to die and damaged
nerves hinder the sensation thus making widespread damage much more likely.
The most common techniques for measuring blood glucose are enzymatic. The
glucose oxidase method is a very popular manual procedure used for self-monitoring.
The Hexokinase method is widely used in laboratories since the procedures for it are carried out by automated equipment.
3.7.1 Glucose oxidase method

The glucose oxidase method is used in a large number of commercially available strip
tests. These simple strip tests allow easy and quick blood glucose measurements. A strip
test product, One Touch II (Lifescan, Milpitas, CA), depends on the glucose oxidase
peroxidase chromogenic reaction. After a drop of blood is combined with reagents on the
test strip, the reaction shown in Eq. (3.15) occurs.
glucoseoxidase

Glucose + 2H 2 O + O 2 GluconicAcid + 2H 2 O 2

(3.15)

Addition of the enzyme peroxidase and o-dianiside, a chromogenic oxygen results in the
formation of a colored compound that can be evaluated visually.
peroxidase

o - dianisine + H 2 O 2
oxidized o - dianisine + H 2 O

(3.16)

Glucose oxidase chemistry in conjunction with reflectance photometry is used to produce

a system for monitoring blood glucose levels (Burtis and Ashwood, 1994). In the Lifescan system (Figure 3.3), a drop of blood is applied to the reagent pad, the strip is inserted
into the meter, and the results appear on a digital display screen 45 s later. The device
automatically starts and times the test and has a range of 0 to 600 mg/dL.
3.7.2 Hexokinase method

Much of the automated equipment is based on the hexokinase method. The general reactions are
hexokinase

Glucose + ATP glucose6phosphate + ADP

(3.17)

G 6 PD

Glucose6phosphate + NAD+ 6phospogluconate + NADH + H+ (3.18)

Glucose is phosphorylated to glucose-6-phosphate in the presence of ATP, Mg2+, and the
enzyme hexokinase. The glucose-6-phosphate formed in Eq. (3.17) is oxidized to 6phosphogluconate by the enzyme glucose-6-phosphate dehydrogenase (G-6-PD) in the

3-12 Chapter 3 Analysis of Molecules in Clinical Medicine

presence of nicotinamide-adenine dinucleotide (NAD+) (Eq. 3.18). The NADH produced
absorbs electromagnetic energy at 340 nm and is proportional to the glucose concentration. This method can be used for glucose detection in both serum and plasma. There are
several other techniques based on the hexokinase method (Burtis and Ashwood, 1994).

LIFESCAN

Front

Back

Apply a drop
of blood to
the strip

The dot turns blue

when enough
blood is applied

109

mg/dL

Sure Step

Figure 3.3 LifeScan, Inc., a system by Johnson and Johnson for self-monitoring glucose levels.

3.8 Amperometric biosensors for oxygen and glucose

3.8.1 Oxygen (PO2)

Amperometry is a technique that measures the amount of current flowing through an

electrochemical cell when a constant potential is applied to the electrodes (Burtis and
Ashwood, 1994). An amperometric sensor that measures the partial pressure of O2 (PO2)
is the Clark electrode. The Clark electrode is actually an electrochemical cell consisting
of two electrodes: a glass covered platinum cathode, and a silver/silver chloride anode.
The silver/silver chloride anode is in a phosphate buffer containing potassium chloride to
buffer the hydroxyl ions produced at the cathode (Eq. (3.20)). Figure 3.4 shows the main
components of the Clark (PO2) electrode.
At the anode, the following reaction occurs
4Ag + 4Cl
4AgCl + 4e

(3.19)

At the cathode, the following reaction occurs:

O 2 + 2H 2 O + 4e
2H 2 O 2 + 4e
4OH

(3.20)

4OH + 4KCl
4KOH + 4Cl
The electrode at which the reductionoxidation (redox) reaction involving the molecule
of interest occurs is often referred to as the working electrode. According to the Nernst

3-13

3.8 Amperometric Biosensors for Oxygen and Glucose

equation (Eq. 3.21), a cell reaction will spontaneously proceed to the right if the potential
of the cell is greater than zero (E > 0) because the concentration of O2 is greater than H20.
A cell reaction will spontaneously proceed to the left if the potential for the cell is less
than zero (E < 0). When the potential for the cell is zero (E = 0), the redox reaction is at
equilibrium.
Platinum cathode

Glass rod

0.7 V
Sample outlet

e
Dissolved
O2

Sample inlet

O2 permeable membrane

+
Cl

Ag/AgCl anode
Phosphate buffer

Ammeter

Figure 3.4 In the PO2 electrode, O2 dissolved in the blood diffuses through a permeable membrane.
Current is proportional to PO2 (Hicks et al., 1987).

E = E0 +

CH O
0.059
2
log
n
CO

(3.21)

where n is the valence of the electrode (cathode) material. E0 is the standard half-cell
potential. It is the half-cell potential when CO = CH 0. A dc voltage between 600 and 800
2
2
mV allows the electron transfer reaction between O2 and H2O to occur. The entire reaction at the working electrode consists of several steps, each of which has its own inherent
rate. These steps include the transport of O2 from the solution to the working electrode,
the transfer of electrons from O2 to H2O, interactions between O2 and H2O (the oxidized
and reduced forms of the species of interest), and the transport of H2O back to the solution. The amount of current produced is controlled by the slowest of these steps. Because
O2 is consumed at the cathode, there is a concentration gradient from the dissolved O2 to
the cathode. It takes about 60 s to establish this concentration gradient for a stable measurement. The cathode may be vibrated to ensure that fresh O2 is available at the cathode.
The current produced by this reaction is directly proportional to the rate of electrons transferred from O2 to H2O. The rate of electrons transferred from O2 to H2O is
directly proportional to the concentration of O2.
3.8.2 Glucose

The first enzyme electrode to measure glucose was developed by Clark and Lyons in
1962 (Figure 3.5). These biosensors still serve as the classical example for measuring

3-14 Chapter 3 Analysis of Molecules in Clinical Medicine

glucose because their operation explains the basic concepts for measuring glucose with
an amperometric enzymatic electrode (Hall, 1991).
Glucose + O2

H2O2

O2

O2

Glucose oxidase

Glucose acid

Pt cathode
0.7 V
Ag/AgCl
anode

Ammeter

Figure 3.5 In the glucose enzyme electrode, when glucose is present, it combines with O2, thus
decreasing the O2 that reaches the cathode.

It is often referred to as a first generation biosensor due to its structure and the
level at which its components are integrated (Taylor and Schultz, 1996). This glucose
sensor uses the PO2 electrode and the enzyme glucose oxidase immobilized on a membrane. The reaction for the basic glucose sensor is
glucoseoxidase

Glucose + O 2 + 2H 2 O Gluconicacid + 2H 2 O 2

(3.22)

Glucose and oxygen react in the presence of the enzyme glucose oxidase. The amount of
O2 consumed, indicated by a decrease in PO2, in the reaction is proportional to the concentration of glucose in the solution. (The concentration of glucose is indicated by the
consumption of O2 by the enzyme glucose oxidase.) The disadvantage of the single oxygen electrode glucose sensor is that it is sensitive to variations of PO2 in the blood.
A dual oxygen electrode glucose sensor (Peura, 1998) eliminates the problem of
variations in PO2 that the single oxygen electrode is sensitive to. As the name implies,
the dual oxygen electrode uses two oxygen electrodes to compensate for the PO2
sensitivity. An enzyme within a gel covers one electrode, thus depleting oxygen when
glucose is present. The other electrode, not covered by the enzyme, senses only oxygen.
The amount of glucose is determined as a function of the difference between the
electrode responses. The disadvantage of enzymatic glucose sensors is that they can only

3.8 Amperometric Biosensors for Oxygen and Glucose

3-15

responses. The disadvantage of enzymatic glucose sensors is that they can only be used in
vivo for a few months and the enzyme is unstable.
A more recent approach to amperometric enzyme glucose electrodes involves
the use of the electron transfer mediator dimethyl ferrocene. Electron transfer mediators,
such as dimethyl ferrocene, are low molecular weight species, which shuttle electrons
between the working electrode and the oxidationreduction center of the enzyme (Taylor
and Schultz, 1996). There are several advantages in using mediators. First, mediators are
much more readily reduced than the substrate co-factor (in this case O2). Second, they
allow glucose measurements that are independent of the variations in PO2 of the sample
(Burtis and Ashwood, 1994). In the series of reactions shown in Figure 3.6, a reduced
form of the mediator dimethyl ferrocene reduces the oxidized form of glucose oxidase.
Measurement using this reaction can be made using whole blood and the current produced is directly proportional to the glucose concentration. One of the disadvantages of
this technique is that dimethyl ferrocene (as well as all mediators) is not very soluble and
ends up adsorbing to the electrode surface.
Electrode surface
e

Dimethyl ferrocene
oxidized form

Glucose oxidase
reduced form

Glucose

Dimethyl ferrocene
reduced form

Glucose oxidase
oxidized form

H 2O 2

Figure 3.6 The sequence of reactions involved in the mediated reaction of a glucose sensor (Taylor
and Schultz, 1996). Dimethyl ferrocene expedites the transfer of electrons from the electrode surface to the redox center of the enzyme glucose oxidase. The use of the mediator dimethyl ferrocene
reduces the sensors dependence on oxygen tension.

Cygnus, Inc. has developed a GlucoseWatch. An electric current is used to extract glucose across the skin into a hydrogel using iontophoresis. Within the hydrogel, the
extracted glucose undergoes a reaction with the enzyme, glucose oxidase to produce gluconic acid and hydrogen peroxide in the presence of oxygen. The hydrogen peroxide then
further diffuses to and reacts on a platinum electrode to produce two electrons, water, and
oxygen (Kurnik et al., 1998).

3.9 Ion-selective electrodes for pH and CO2

Ion selective electrodes are used to measure the activity of a specific ion in a cell or solution. An ion selective electrode contains a membrane that is permeable to a specific
cation or anion. In the presence of the specific ion, a membrane potential results which
indicates the amount of ion activity. pH is the negative logarithm to the base 10 of the
hydrogen ion concentration [H+] in mol/L.

3-16 Chapter 3 Analysis of Molecules in Clinical Medicine

pH = log10[H+]

(3.23)

The measurement of pH is obtained by using an ion selective electrode (ISE) with a glass
membrane. To prevent pH measurement errors that occur due to other ions, a highly selective H+ glass membrane is desired. One such glass composite consists of silicon dioxide, lithium oxide, and calcium oxide in the ratio 68:25:7 (Burtis and Ashwood, 1994).
1 | glass membrane | 2

(3.24)

The glass membrane separates the test solution on side 2 from the reference solution, usually hydrochloric acid of known pH, on side 1 of Eq. (3.24). When the pH of
the solutions on side 1 and 2 differ, a potential across the membrane develops, V = V1
V2. The potentials of solution 1 (hydrochloric acid) can be recorded with a reference electrode such as a silver/silver chloride electrode. The potential of solution 2 can be recorded with a calomel electrode. Ideally, the pH electrode obeys the Nernst equation

E = K + (2.303

RT
) log(a i)
ZF

(3.25)

where E is the potential, R the universal gas constant, F is the Faraday constant (= 96487
C/mol, C = coulomb), T the absolute temperature, Z the ionic charge, ai the activity of the
ion, and K is a constant containing contributions from various sources. The activity of an
ion, ai, equals the activity coefficient i times its concentration c.
ai = i c

(3.26)

Factors such as other ions in the solution and the strength of H+ in the solution
influence i. i decreases as c increases. i for plasma is about 0.75 for Na+, 0.74 for K+,
and 0.31 for Ca2+. In solutions where the primary ion (H+) is of very low concentration
(M), the membrane potential is not directly proportional to the logarithm of the activity
of the diffusible ion H+ in solution 1 or 2. The relationship is better expressed by the NikolskiEisenman equation.
Or the potential can also be calculated from
E = 0.0615 log10

[H + ]i

[ H + ]o

(3.27)

where [H+]i is the concentration of H+ inside the cell and [H+]o is the concentration of H+
ions outside the cell.
IQ Scientific Instruments (2001) offers pH measurement without glass. A
stainless steel probe has a silicon chip sensor and can be stored dry. The ISFET consists
of a silicon semiconductor substrate with two electrical contacts (source and drain) a
small distance apart. Deposited on the substrate between the source and drain is a silicon
electrical insulator. Hydrogen ions at or near the surface of the insulator cause a variable
voltage potential between the insulator and the underlying semiconductor material be-

3.9 Ion-Selective Electrodes for PH and CO2

3-17

tween the source and drain. The variable voltage potential is proportional to the relative
concentration of hydrogen ions in the sample solution. The pH can then be derived from
this voltage to a very high level of accuracy.
The measurement pH is the measure of the concentration of H+ protons. The pH
level in the body is more strictly regulated than glucose. The normal range of pH is 7.36
to 7.44. A pH of less than 7.36 is called acidosis. If the pH is less than 7.0, a person will
go into a coma and die. A pH greater than 7.44 is called alkalosis, and a pH greater than
7.8 causes the bodys muscles to seize up, a condition known as tetany. The explanations
for these phenomena lie in acidbase chemistry. Molecules in the body, particularly proteins, have sites where protons can associate or dissociate, and when the bodys pH is
altered, the charge of that particular portion of the molecule is essentially changed. The
electrostatic charges on the parts of the molecule may attract or repel in different ways
than they should. As a result, proteins do not fold properly and DNA does not form as it
should, among other things.
The metabolism of substances in the body tends to produce protons. With the
exception of vomiting, protons leave the body via the kidney. The kidney can excrete
acids directly or it can produce bicarbonate, which combines with a proton to form carbonic acid. Carbonic acid spontaneously breaks down into water and carbon dioxide.
Another way the body eliminates acid is by exhaling carbon dioxide through the lungs.
An individual can exhale too much or too little carbon dioxide, thus changing the acidity
of the blood. Or, if the kidney is prevented from eliminating acid, acid buildup can occur,
causing a condition known as acidosis. Both the lungs and the kidneys regulate the
bodys pH level. Having two ways to eliminate acid allows one method to compensate if
the other becomes impaired. If the pH level is abnormal, there are three things that can be
measured: carbon dioxide concentration, pH, and bicarbonate concentration. By measuring these three concentrations, we can determine if the kidneys or lungs are functioning
abnormally. If the problem is abnormal CO2 concentration, then we know the problem is
with the lungs. If the bicarbonate concentration is abnormal, then a condition called
metabolic acidosis exists and the cause might be either the kidney or something within
the metabolic pathway. Further studies are required for abnormal bicarbonate concentration such as sodium concentration or chloride concentration measurements of the blood.
In the CO2 electrode, blood comes in contact with a plastic membrane. Holes in
the membrane permit gas to diffuse through, but block everything else. The CO2 diffuses
to a chamber containing a pH electrode, forms carbonic acid, and increases the acidity.
The reaction for this process is
CO2 + H2O H2CO3 H+ + HCO3
Since CO2 concentration is proportional to H+ concentration; measurements from the pH
electrode yield PCO2.

3.10 Flame photometry

Flame photometry, a subset of emission photometry, is a technique commonly used to
quantify sodium, potassium, and lithium concentrations in bodily fluids. It is a suitable

3-18 Chapter 3 Analysis of Molecules in Clinical Medicine

technique for quantifying these elements, because compared to calcium and magnesium,
they have low excitation potentials. Figure 3.7 shows a block diagram of a flame photometer. The purpose of the flame is to provide energy to the elements such that their
valence electrons will be excited from ground states to higher excited states. Electrons in
the excited states are unstable and emit energy as photons of a particular wavelength as
they return to their ground states. The wavelength(s) of the light emitted by an element is
characteristic of that element. When sodium, potassium, and lithium are present as
cations in an aqueous solution and exposed to a flame, sodium produces yellow, potassium violet, and lithium red colors. The intensity of the light given off is directly proportional to the number of photons being emitted, which in turn is directly proportional to
the number of atoms, or concentration of cations, in the solution.
Flame

Lens

Filter

Collimating lens

Detector Readout
Fuel inlet
Capillary

Fuel inlet
Oxygen inlet

Aqueous sample

Figure 3.7 The flame photometer aspirates a sample containing metal ions and heats it to incandescence. Detector output is proportional to concentration.

In flame photometry, the monochromator and detector are similar to those found
in the spectrophotometer, although the detector is measuring emission as opposed to absorption. The distinguishing features of the system are the flame and the atomizer. The
atomizer draws the sample solution containing the cations through an aspirator and into
the flame. This is achieved by passing a gas over the upper outlet of a capillary tube at a
high velocity, while the lower end of the capillary tube is inserted into the sample. The
fuel used to generate the flame can be either propane, natural gas, or acetylene mixed
with compressed air or oxygen. Lithium can be used in calibration because it does not
naturally occur in the body (provided the patient is not taking lithium medication). A
known concentration of lithium is added to the sample. Then a ratiometric technique can
be used to compare the ratio of sodium and potassium to lithium.

3.10 Flame Photometry

3-19

3.10.1 Lithium

Lithium is not produced in the body and does not naturally appear in the body. The purpose of lithium measurement is to measure lithium carbonate, which is used in the treatment of the psychiatric disorder manic depression (also known as bipolar disorder).
3.10.2 Sodium and potassium

Sodium and potassium are used by the body to maintain concentration gradients in nerve
and muscle cells, which allows them to conduct action potentials. To keep these cells
working properly, the amount of sodium and potassium in the body must be regulated. It
is the responsibility of individual cells to do the regulation. Sodium and potassium enter
the body via eating. For people who do not eat properly, it is the kidneys job to regulate
and compensate the levels of sodium and potassium after eating. For example, if an individual eats a lot of salt, the excess salt in their blood will be passed in the urine. Potassium chloride can be injected into the blood stream to raise the level of potassium in the
body, although it is important to be extremely careful with the amount that is injected.
Too much potassium can kill a person by stopping their heart as a result of decreasing the
concentration gradient, and hence the ability to generate an action potential in the heart
muscle cells.

3.11 Mass spectrometry

Mass spectrometry is a powerful analytical technique that is used to identify unknown
compounds, quantify known materials, and to elucidate the chemical and physical properties of molecules. Mass spectrometry can provide valuable information to a wide range of
professionals including physicians, biologists, chemists, and engineers. A mass spectrometer separates a material according to its atomic or molecular mass and the sample of
molecules can be a gas, liquid, or solid, as long as the specimen is not too volatile. Mass
spectrometry involves ionization of a gas-phase sample, fragmentation of the ionized
species, sorting of the fragments according to their mass and charge, and measurement of
ion abundance at each mass. As far as the scope of mass spectrometry is concerned, many
different techniques and types of instrumentation have been devised to accomplish each
of the above processes. This discussion will be limited to those most useful in the toxicology laboratory.
3.11.1 Sample inlet

Figure 3.8 shows that the specimen containing known or unknown molecules enters the
mass spectrometer through the sample inlet, where it is vaporized under a high vacuum.
Several types of sample inlet systems exist that accommodate certain aspects of the sample such as its temperature, vapor pressure, and its volatility. An important inlet system is

3-20 Chapter 3 Analysis of Molecules in Clinical Medicine

a gas chromatograph interface, which works by allowing the components of a mixture to
enter the mass spectrometer continuously as they emerge from the chromatograph.
3.11.2 Ion source

Figure 3.9 shows the ion source where the sample gets ionized. Many different methods
to ionize the sample exist, one of which is the heated filament. A filament, (rhenium or
tungsten) when heated to greater than 2000 C by electric current, emits electrons under
the influence of an electric field. Typical current flow is about 500 A. The electrons are
focused in a magnetic field, which causes them to spiral toward the positively charged
target. Electron energy is typically 70 to 80 eV. When a sample is bombarded by the
high-energy electrons, the ionization process is initiated.

Sample inlet
Sampling
device

Interface

Porous plug

Ion
source

Mass
analyzer

Vacuum
system

Detector

ADC

Figure 3.8 The mass spectrometer separates molecules in a high vacuum after they have passed
through a porous plug.

By bombarding molecules of the vaporized sample with electrons from a heated

filament, electrons can be removed from the molecules. When an electron is removed
from a molecule, it becomes positively charged or ionized (+). Since the mass of an electron is negligible relative to the mass of a molecule, the mass of the molecular ion remains essentially the same
ABC + e ABC+ + 2e
ABC + e ABC

(3.28)

3.11 Mass Spectrometry

Pole of magnet
Filament

3-21

Focusing elements
(electric field)
N

Repeller

Electron trap
(positively
charged target)

To mass
analyzer

S
Pole of magnet

Figure 3.9 An electron impact ion source bombards the molecules and carrier gas from a gas
chromatography column with electrons. The electron collisions with ion ABC (Eq. (3.28)) produce
both positive and negative ions. The positive ions are directed by focusing elements to the mass
analyzer.

For a time span of 0.1 to 10 ns after the collision of the electron with the ion,
several of these molecular ions disintegrate, forming radicals () and positively charged
fragment ions.
ABC+ AB+ + C +
A + + BC +
AB+ + C (Loss of neutral)
+

(3.29)

AC + B (Rearrangement)
etc.
Not all of the positively charged molecular ions undergo fragmentation. These
unfragmented positively charged molecular ions are drawn out of the ion source by an
electric field while the vacuum system pumps the negative and neutral fragments away.
An electric field then appropriately focuses and accelerates them into the mass analyzer.
The intensity of the electron beam and size of the sample determine the number of fragments and molecular ions, whereas the chemical stability of the molecule determines the
relative amounts of molecular ions and fragment ions. Figure 3.10 is an example of a
mass spectrum for a sample and its resulting parts. A mass spectrum is a plot that represents the relative ion abundance at each value of m/z (mass to charge ratio).

Sample abundance

3-22 Chapter 3 Analysis of Molecules in Clinical Medicine

AB+
C+
ABC+

A+

AC+
m/z

Figure 3.10 The mass spectrum displays the relative abundance of charged molecular ions and
fragments.

3.11.3 Mass analyzer

The function of the mass analyzer is to separate the stream of ions produced from the ion
source by their mass to charge ratio (m/z). This is usually done with a magnetic field.
Figure 3.11 shows the main components of a quadrupole mass analyzer.

Detector

+
Ion
source

RF voltage
supply

DC voltage
supply

Figure 3.11 A quadrupole mass analyzer accelerates ions from the ion source. The dc and RF voltages that are applied to the four ceramic rods stabilize the paths of ions with selected m/z values.
Ions that do not stabilize collide with the rods and do not reach the detector.

3.11.4 Detector

The three most commonly used detection schemes are the electron multiplier, photographic emulsions, and Faraday cups (Haven et al., 1995). Figure 3.12 shows the main
components of an electron multiplier. Positively charged ions collide violently into the

3.11 Mass Spectrometry

3-23

cathode, thus releasing secondary electrons. These secondary electrons are attracted to
the first dynode, which is 100 V more positive than the cathode. A larger number of tertiary electrons are released and attracted to the second dynode, etc. to build up a current of
about 1 A, which is measured at the anode.
Dynodes
Cathode

Anode

Horn

Collector

Ion+

Dynodes
Quadrupole
mass analyzer
rods
X-ray shield

Figure 3.12 Block diagram of an electron multiplier. An electron multiplier is sensitive enough to
detect the presence of a single ion. A 10 to 12-dynode electron multiplier produces a signal amplification of 105 to 106. (Haven et al., 1995).

3.12 Carbon dioxide concentration measurement by infrared transmission spectroscopy

A spectrum usually shows the intensity of electromagnetic waves as a function of wavelength. The intensities depend on the medium(s) the waves pass through. Spectroscopy is
the analysis of spectra (components of electromagnetic radiation). Spectra are formed by
a spectroscopean optical instrument that physically breaks up electromagnetic radiation
into its component parts, usually with a prism or grating. Infrared transmission spectroscopy is the analysis of spectra over the range of wavelengths in the infrared region (770
to 1000 nm) of the electromagnetic spectrum and can be used to measure the concentration of carbon dioxide in a gaseous mixture.
The majority of carbon dioxide in the body is expired through the lungs as an
end product of the respiration cycle. Measuring the amount of CO2 expired by the lungs
can be used to help diagnose disease. Gases with dipole moments such as CO2, N2O, or
water vapor (H2O), absorb a specific wavelength of light in the infrared region of the
spectrum. Gases without dipole moments such as N2 and O2 do not absorb any power
(infrared radiation). Carbon dioxide concentration can be measured by passing a particular wavelength of infrared light through a chamber containing a sample of carbon dioxide

3-24 Chapter 3 Analysis of Molecules in Clinical Medicine

and noting the power of the infrared light transmitted to the sample, and the power of the
infrared light transmitted through the sample. Using infrared light of a particular wavelength characteristic for measuring CO2, the concentration of CO2 in a mixture of gases
can be determined through Beers Law
Pt = P0e aLc

(3.30)

where Pt is the power per unit area of transmitted infrared light received by the detector,
P0 is the power per unit area of infrared light entering the sample, a is the absorption coefficient, L is the path length of light through the gas, and c is the concentration of gas.
Figure 3.13 shows the basic components of an infrared spectroscopy system (Primiano,
1998). The infrared source is modulated by rotating vanes similar to a windmill. The detector compares transmission through the sample cell with transmission through a reference cell.

Infrared light source

Sample

Modulator

Sample
cell

Detector

Signal processor
and display

Figure 3.13 An infrared transmission spectroscope measures the absorption of infrared light by a
sample drawn through the sample cell by a vacuum pump.

3.13 Nitrogen by emission spectrometry

The concentration of nitrogen, N2, in a sample of a mixture of gases can be measured by
emission spectrometry. Nitrogen concentration in a mixture of gases is a measurement
made in respiratory and pulmonary medicine. Emission spectrometry involves applying a
voltage of 600 to 1500 V dc between two electrodes on opposite sides of an ionization
chamber in which the pressure is low (1 to 4 mm Hg or 150 to 550 Pa). Under these conditions, the nitrogen in the chamber is ionized and emits ultraviolet light (wavelengths
from 310 to 480 nm). Figure 3.14 shows the main components of a nitrogen analyzer
(Primiano, 1998).

25

3.13 Nitrogen by Emission Spectrometry

Gas mixture conduit
Needle valve
Phototube

Sample inlet tube

Electrodes
Reflecting surface

Amplifier
and signal
processing

Constant voltage
power supply

Ionization
chamber

vo

Optical filter

To vacuum pump

Figure 3.14 Emission spectroscopy measures ultraviolet light emitted by ionized nitrogen
(Primiano, 1998).

3.14 Drugs by fluorometry and chromatography

3.14.1 Fluorometry

Fluorescence is the emission of energy in the form of light as a result of the electrons of a
molecule returning to their ground state. When certain molecules absorb energy (in the
form of electromagnetic radiation) their electrons are raised to higher energy levels. As
the electrons of the molecule return to their ground states, they fluoresce. A fluorometer
is an instrument that measures the intensity of light produced when the electrons of a
molecule return to their ground state. Thus, fluorometry is defined as measuring the relationship between the concentration of a substance and the intensity of the fluorescence
produced by that compound when it is excited by radiation. Fluorometry is used to measure therapeutic drugs such as phenobarbital (treatment of epilepsy), enzymes such as proteases (used to break down proteins), hormones such as cortisol, and analytes such as
catecholamines (e.g. epinepherine and norepinepherine) and bilirubin. Figure 3.15 shows
the components of a fluorometer, which include an excitation source, a primary (excitation) filter, a secondary (emission) filter, a cuvette sample, a detector, and a readout device.
Although the components of a fluorometer are basically the same as those of a
spectrophotometer, fluorometry is up to four orders of magnitude more sensitive than
spectrophotometry (Wheeler, 1998). In spectrophotometry, the concentration of the sub-

3-26 Chapter 3 Analysis of Molecules in Clinical Medicine

stance is determined by the difference of the absorbance between a solution that has zero
absorbance (the solution used to calibrate the spectrophotometer) and the absorbance of
the unknown solution. When the absorbance of the unknown solution is relatively small,
small errors in the measurement of the zero absorbance solution can cause relatively large
errors in the final determination. Fluorometry avoids this problem by measuring fluorescence directly from the sample without reference to another measurement. Though a disadvantage of fluorometry is that measurements are also susceptible to inner-filter effects.
These effects include excessive absorption of the excitation radiation (prefilter effect) and
self-absorption of atomic resonance fluorescence (postfilter effect).
Primary filter

Cuvette

I0

If

Secondary filter
Detector
Readout

Figure 3.15 Block diagram of a fluorometer. The primary filter passes only wavelengths that excite
the fluorescent molecule. The secondary filter blocks all scattered excitation wavelengths and
passes only the scattered fluorescent wavelengths. The secondary filter and detector are at a right
angle to the primary beam in order to avoid direct transmission of the light source through the sample to the detector.

Theory of fluorometry

The theory behind fluorometry can be derived from the BeerLambert law
where

I = I010aLc

I = intensity of transmitted light

I0 = intensity of the incident light
a = absorptivity of the sample
We rearrange to determine the amount of light absorbed:

(3.31)

3.14 Drugs by Fluorometry and Chromatography

3-27

I0 I = I0(1 10aLc)
The fluorescence intensity, IF, is proportional to the amount of light absorbed,
I0(1 10aLc), and the fluorescence efficiency, , is the ratio of light emitted to light
absorbed, giving
IF = I0(1 10aLc)

(3.32)

We can expand through a Taylors series, rearrange, and convert the logarithm base to
yield
IF = I0 (2.23aLc)
Thus the fluorescence intensity is directly proportional to the concentration of the fluorescent molecules in the sample and the excitation intensity.
Fluorometry is limited by several factors, including concentration of the sample.
For more details, see (Burtis and Ashwood, 1994).
Spectrofluorometers replace the fixed secondary filter with an emission monochromator and sweep it at 5 to 50 nm increments to yield a spectrum of emission magnitude versus wavelength. Some spectrofluorometers also sweep an excitation monochromator instead of having a fixed primary filter. In time-resolved fluorometers, long-lived
fluorescent molecules are excited by pulsed light and their decay is monitored over 0.6 to
100 s. Laser-induced fluorometers measure emission from individual fluorescent or
fluorescent-labeled molecules, cells, or particles passing through a flow cell cuvette.
3.14.2 Chromatography

Chromatography is an efficient technique used to separate components in a mixture.

Some components include drugs such as antibiotics, therapeutic drugs, or abused drugs.
There are two main phases in a chromatographic system: a stationary phase and
a mobile phase. Chromatographic separation relies on the components in a mixture being
separated by the stationary phase. The basis for chromatography is the difference in affinity for the stationary phase versus the affinity for the mobile phase of the components in a
mixture. The greater the affinity for the stationary phase, the less a component will move.
Likewise, greater affinity for the mobile phase indicates that the component will migrate
farther. In order for separation to occur, the mobile phase must move past the stationary
phase. In gas chromatography, the mobile phase is the gas and the stationary phase is the
solvent, which covers a solid support in the column. The components in a chemical mixture react differently with each of the phases in the column. For example, as the component mixture is injected into the column, some components interact with the stationary
phase longer than others. As a result, the components in the mixture become separated in
the column. The components that interacted for a shorter time with the stationary phase
get washed out of the column by the solvent more quickly than the strongly interacting
components.

3-28 Chapter 3 Analysis of Molecules in Clinical Medicine

A chromatogram is a plot of the relative amounts of the components in the mixture as a function of the amount of time it takes for them to get washed (eluted) off the
column or pass through the column. The following information can be obtained from a
chromatogram. The number of peaks plotted indicates the complexity of the mixture.
Comparing the position of the peaks relative to the position of standard peaks can provide
qualitative information on the sample composition and also reveal the performance of the
column. Measuring the area under each peak yields a quantitative evaluation of the relative concentrations of the components.
3.14.3 Gas chromatography

Gas chromatography (GC) is a type of chromatography that uses an inert gas as the mobile phase. GC is used to separate compounds that are volatile at the temperature in the
column of the chromatograph. These mixtures of volatile organic compounds, often small
organic molecules such as therapeutic drugs, are measured from a sample of urine or
sometimes blood. Figure 3.16 shows a gas chromatograph consisting of several components: a carrier gas, a column, a column oven, an injector, a detector, a recorder, and a
flowmeter.
Column

Column oven

Injector
Flow meter

Detector

Flow control

Carrier gas

Figure 3.16 In a gas chromatography system, the sample is injected into a carrier gas and flows
through a long column to the detector.

There are several advantages of GC: it is fast, extremely sensitive, and requires a very
small sample.
Carrier gas

The inert carrier gas, such as argon, helium, or nitrogen, is the mobile phase in the system
because it moves the components in the evaporated solute through the column. The carrier gas has very little effect on the rate of retention of components in the solute.

3.14 Drugs by Fluorometry and Chromatography

3-29

Column

The column in GC is usually enclosed in an oven with computerized temperature regulation. The ovens are usually programmed to produce a linear increase in temperature over
a span of time, depending on the solute. When the column is heated over a range of temperatures, components in the solute with lower boiling points will elute off the column
first, followed by components with higher boiling points. Thus, heating the column
gradually allows the components to be separated.
The column (stationary phase) separates the components in the sample because
they each have different retention times. The two most common types of columns used in
GC are packed columns and capillary columns. Packed columns are typically between 1
and 4 mm in diameter and 1 to 4 m in length (Burtis and Ashwood, 1994) They are
packed with a solid support material such as diatomaceous earth (a type of algae with cell
walls containing Ca++) and then coated with a liquid (the stationary phase). Capillary
columns are typically between 0.2 and 0.5 mm in diameter and 10 to 150 m in length
(Burtis and Ashwood, 1994). In capillary columns, the inner wall of the tube is coated
with the stationary phase.
Both types of columns are made with stainless steel or glass. The advantage of
glass is that it is inert and the advantage of stainless steel is that oxide films do not develop on the surface, thus preventing reactions with solutes. The advantage of capillary
columns is that they can detect very small amounts of a component and are highly efficient. The advantage of packed columns is that they can handle a large sample size and
require less maintenance.

Detector

The purpose of the detector is to quantify the components in a sample. It does this by
providing an electric signal proportional to the amount of the component coming off the
column. Some of the commonly used detectors in GC include the flame ionization detector, the thermal conductivity detector, the electron capture detector, the nitrogen phosphorus detector, and detectors based on mass spectroscopy. The combination of mass spectrometry and GC is ubiquitous in clinical chemistry laboratories in hospitals and is a very
powerful tool.

Resolution

Figure 3.17 shows that the detector yields a chromatogram. For a flow rate F, the peak of
a solute, A, appears after retention time tr. The volume that has passed before the peak is
the peak volume and is given by Vr(A) = trF. Band spreading will broaden the peak to a
Gaussian shape. An efficient column minimizes band spreading. The degree of efficiency
is the number of theoretical plates, N = [Vr(A)/s(A)]2, where s(A) is the standard deviation of the peak volume. The width of a Gaussian peak at 1/2 its height, w1/2, is 2.354
standard deviations, from which N = 5.54[Vr(A)/w1/2(A)]2.

3-30 Chapter 3 Analysis of Molecules in Clinical Medicine

w 1/2
INJ

tr

1/2
height

Figure 3.17 In chromatography, the peak appears after retention time tr.

The resolution of two solute bands is a measure of their degree of separation in

terms of relative migration rates and bandwidths and is
Rs =

Vr (B) Vr (A)
[ w(A) + w(B)]/2

(3.33)

where w(A) and w(B) are the peak widths measured at the corresponding base. Inadequate separations occur for Rs < 0.8 and baseline separation occurs for Rs > 1.25 (Bowers
et al., 1994).
Example 3.1 For a gas chromatograph with a flow rate of 1.5 L/min, solute A has a
retention time of 1 min and solute B has a retention time of 2.5 min. If the peak width for
solute A is 0.5 L and the peak width for solute B is 1.5 L, is the resolution sufficient?
Since the flow rate of the solutes and peak widths are known, we only need to
find the peak volume before we can use Eq. (3.33) to determine if the resolution is sufficient.

w(A) = 0.5 L
w(B) = 1.5 L
Vr (A) = t rA F = 1 min 1.5 L/min = 1.5 L
Vr (B) = t rB F = 2.5 min 1.5 L/min = 3.75 L
Rs =

Vr (B) Vr (A)
3.75 L 1.5 L
=
= 2.25
[ w(A) + w(B)]/2 [0.5 L + 1.5 L] 2

Since Rs > 1.25, the base separation of the bands is sufficient.

3.14 Drugs by Fluorometry and Chromatography

3-31

Coupling GC to MS

GC separates mixtures of volatile organic compounds. Combining the gas chromatograph

with the mass spectrometer (MS) is a powerful tool. GC has high separation efficiency
while mass spectrometry offers detailed structural information and conclusive identification of molecules when corresponding reference spectra are available.
When interfacing gas chromatography with mass spectrometry, several compatibility issues arise. Although both are gas phase microanalytical techniques, gas chromatography operates at or above atmospheric pressure whereas mass spectrometry requires a high vacuum. In most cases, an interface is required to minimize these incompatibilities. For a capillary column, there is a direct interface and open split interface. For
a packed column, the jet separator is commonly used. For the capillary-direct interface,
the column is directly coupled to the MS ion source. The advantage of the direct interface
is total inertness from the head of column to the ion source. The disadvantages are that
chromatography is affected by the mass spectrometers vacuum (i.e. a shorter column is
needed), venting the MS vacuum system to change the column may be required, and it is
limited to > 12.5 m narrow bore (0.2 mm) and > 25 m wide bore (0.3 mm) columns.
The jet separator is the most efficient GC/MS interface. It can remove about
90% of a light carrier gas such as helium, yet pass all but a few percent of the heavier
sample compounds. This is accomplished by capitalizing upon differences in gas diffusion rates in a supersonic jet stream. As the gas from the GC passes through a nozzle expanding into an evacuated region, the momentum of the heavier sample gas is increased
relative to the carrier gas. The ratio of sample to carrier gas molecules in the region of the
jet stream increases, enriching the sample passing through the nozzle leading to the mass
spectrometer. The jet separator operates with maximum efficiency at column flows of 25
mL/min and is capable of transmitting a wide range of molecular weights from permanent gases to high molecular weight compounds. Efficiency improves as mass (m/e) or
momentum (mv) of the sample gas increases relative to the carrier.
3.14.4 Liquid chromatography

Liquid chromatography (LC) is a type of chromatography that uses a liquid of low viscosity as the mobile phase. The stationary phase through which this liquid flows is an
immiscible liquid coated on either a porous support or absorbent support. One commonly
used type of liquid chromatography is high-performance liquid chromatography (HPLC).
HPLC is used to separate components dissolved in solution by conducting the separation
process at a high velocity and with a pressure drop. Figure 3.18 shows a schematic of an
HPLC instrument, indicating the major components: the reservoir of mobile phase, a
high-pressure pump, the sample injection port, a separation column, and a detector.
An HPLC instrument operates by injecting a liquid sample via a syringe into the
stream of mobile phase that is being pumped into the column. A detector records the
presence of a substance in the effluent. The detector can observe changes in refractive
index, UV light absorption and fluorescence. Just as in gas chromatography, a mass spectrometer can be coordinated with the HPLC to help identify the components of the mixture.

3-32 Chapter 3 Analysis of Molecules in Clinical Medicine

Flow
splitter

Vacuum
pump
Solvent 1

Sample
injection port

Solvent 2

Degasser 1

Pressure
gage

Degasser 2
Mixing
vessel

Pre-column
Differential
detector

High pressure
pump

To waste

To waste or
fraction collector

Figure 3.18 A schematic diagram of a high-performance liquid chromatography instrument.

3.14.5 Ion exchange chromatography

Another type of chromatography that is widely used in the separation and purification of
biological material is ion exchange chromatography. Ion exchange chromatography operates on the principle that charged molecules in a liquid phase pass through a column
where the stationary phase has a charge opposite that of the molecules in the liquid. Due
to the high affinity of charged molecules for opposite charges, this method is highly selective. It requires a solution with a specific pH or ionic strength before molecules will
elute from the column. Ion exchange chromatography is used in the separation of both
proteins and amino acids.

3.15 Electrophoresis
Electrophoresis is a technique used to separate charged molecules in a liquid medium
with an electric field. It is extensively used in separation of serum proteins, separation of
proteins in urine, determination of molecular weight of proteins, DNA sequencing, genetic disease diagnosis, and comparison of DNA sequences in forensic testing.
3.15.1 Components of an electrophoretic system

An electrophoretic system consists of a power supply, a chamber containing a support

medium, and an anode and cathode made of metal, such as platinum. Carbon can also be
used for the anode and cathode (Figure 3.19). Under the influence of an electric field,
negatively charged anions in a liquid medium migrate to the anode (+). Positively
charged cations migrate to the cathode (). Velocity depends on their charge. Several

3.15 Electrophoresis

3-33

factors influence the distance and rate of migration. Electrophoresis works because when
a charged molecule is placed in an electric field, the molecule sustains a force, which is
proportional to the charge of the molecule and the strength of the electric field.

Cover

Anode
Cathode

Insulating plate

+
Power
supply

Buffer

Electrophoresis support medium

Figure 3.19 In an electrophoresis system, charged molecules move through a support medium because of forces exerted by an electric field.

Power supply

The power supply provides a constant voltage, current, or power across the support medium. Most power supplies for electrophoresis systems provide between 50 and 200 V.
The heat introduced into the electrophoretic system because of the resistance of the support medium is
Heat (energy) = Pt = VIt

(3.34)

where V is voltage, I is current, and t is time. There is a difference between an electrophoretic system that uses a constant voltage supply and one that uses a constant current supply. In a system with a constant voltage supply, the migration rate increases with time
because the resultant current generates heat, which results in thermal agitation of the dissolved ions. This causes a decrease in the resistance of the medium, which further increases the current. Increasing the current generates more heat and water starts to evaporate more quickly. Water loss from the system causes an increase in the ion concentration
of the medium, which also lowers the resistance and thus increases the migration rate of
the different solutes. Migration rates that are too fast or that change with time are undesirable because of decreased resolution of the solutes in the gel. Because of this, it is
more desirable to have a constant current through the gel than a constant voltage. Presentday systems regulate the voltage in an attempt to maintain constant current. Other systems regulate temperature in order to prevent the gel from melting.

3-34 Chapter 3 Analysis of Molecules in Clinical Medicine

Support medium

One of the most commonly used support mediums is agar. Agar is so prevalent in clinical
labs that electrophoresis is often referred to as agarose gel electrophoresis. Clinically,
electrophoresis is performed on a plastic support covered with a thin (1 mm) layer of agarose gel. Agarose has the following advantages: after electrophoresis is performed and
the plate is dried, the agarose is very clear, which enables a densitometer to easily examine it; agarose absorbs very little water because it contains very few ionizable groups.
Another commonly used support medium is cellulose acetate. Cellulose acetate
membranes are produced by reacting acetic anhydride with cellulose. These membranes
consist of about 80% air in pockets in-between interconnected cellulose acetate fibers.
These air pockets fill with liquid when the membranes are placed in a buffer. Cellulose
acetate electrophoresis (CAE) has the advantage of being relatively quick (20 to 60 min).
CAE membranes can be made clear for densitometry by using a solution that dissolves
the cellulose acetate fibers.
Electrodes and chamber

Electrodes may be made of metal such as platinum, or of carbon. Gel trays may be made
from UV transparent acrylic to enable direct observation of the migration of different
molecules.
3.15.2 Electrophoretic mobility

There are several factors that cause different migration rates among the molecules in a
mixture. They are: the net charge of the molecule, q; the size, shape and mass of the
molecule; the strength of the electric field, E; the viscosity of the medium; the temperature of the medium. A molecule in the medium experiences a force FE = qE in the direction of the electric field that is proportional to the strength of the electric field, E, and the
net charge of the molecule, q. The molecule also experiences a drag force FD = fv =
6rv in the opposite direction of movement which is proportional to the molecules resistance f, velocity, v, the viscosity, , of the buffer solution in which it is migrating, and
the ionic radius, r, of the solute molecule, Q.
FE = FD

(3.35)

qE = fv
vss =

qE
f

(3.36)

The mobility of a charged particle is defined as its steady state velocity divided by the
strength of the electric field.

3.15 Electrophoresis

electrophoretic mobility =

vss = q
E

3-35

(3.37)

One mobility unit is defined as 105 cm2/(Vs).

Analysis and quantification

Electrophoresis plates are usually dried before analysis to prevent diffusion of the migrated solutes. Most proteins, nucleic acids, and other molecules require staining in order
to be seen. There are cameras available that can photograph the electrophoresis gel, using
filters specific for different stains. A commonly used stain for visualizing DNA is
ethidium bromide, which is fluorescent under a light known as a Woods lamp. The bands
appearing in the gel, or on the plate, can be identified by inspection under the Woods
lamp, but a densitometer is required for accurate quantification.
Densitometry is commonly used in clinical laboratories. A densitometer measures the transmittance of light through a solid electrophoretic sample, similar to how a
spectrophotometer operates. An electrophoresis plate is placed over the exit slit of the
monochromator of the densitometer. The plate or sample is then slowly moved past the
exit slit while the amount of transmitted light is measured. As the sample specimen is
moved past the exit slit, the location and absorbance of each band of the electrophoretic
sample are recorded on a chart.
Serum protein electrophoresis

Electrophoresis is used to separate the proteins in blood serum. Figures 3.20 and 3.21
show that the first peak is albumin. It is produced in the liver and performs many tasks.
For example, it thickens the blood, binds steroids, and binds carbon dioxide. The other
three peaks are referred to as alpha, beta, and gamma. The proteins that make up gamma
are referred to as gamma globulins. The most prevalent gamma globulins are antibodies
(the term antibodies and immunoglobulins are synonymous since antibodies are part of
the immune system). Various diseases can be diagnosed with the information contained
in these peaks. For example, if the albumin is low, that can be a sign of liver disease. The
gamma globulin peak in Figure 3.21 is a spread of many types of antibodies as a result of
a B-lymphocyte tumor (bone tumor).

3.16 DNA sequencing

There are 20 standard amino acids that make up all proteins. All have a carboxyl group
and an amino group bonded to an carbon. They are distinguished by their different side
chains (R groups). They can be separated, identified, and quantified by ion-exchange
chromatography. In high-performance liquid chromatography (HPLC), 0.4 m spherical
cation-exchange resin is packed into a cylindrical column about 5 mm in diameter and

3-36 Chapter 3 Analysis of Molecules in Clinical Medicine

200 mm long. Amino acids in a buffer solution are pumped through the column at about
13 MPa for about 40 min. Amino acids with the least positive charge bind weakly to the
resin, move most rapidly through the column and elute first to be spectrophotometrically
detected at wavelengths less than 300 nm. Amino acids with more positive charge bind
more tightly to the resin and elute later. The area under the peak of each amino acid on
the chromatogram is proportional to the amount of that amino acid (Lehninger et al.,
1993).
Albumin

Migration distance

Figure 3.20 This serum protein electrophoresis demonstrates a normal pattern, with the largest
peak for albumin.

Albumin

Migration distance

Figure 3.21 This serum protein electrophoresis demonstrates a decrease in the albumin and an
increase in gamma globulins.

Amino acids can be covalently joined by peptide bonds to form polypeptides,

which generally have molecular weights less than 10,000. Peptides can be obtained by
purification from tissue, by genetic engineering, or by direct chemical synthesis. Within

3.16 DNA Sequencing

3-37

the cell, amino acids are synthesized into a longer polypeptide sequence (a protein)
through the translation of information encoded in messenger RNA by an RNAprotein
complex called a ribosome.
To separate and purify proteins, cells are broken open to yield a crude extract.
Differential centrifugation may yield subcellular fractions. Ion-exchange chromatography
can separate proteins by charge in the same way it separates amino acids. Size-exclusion
chromatography separates by size. Affinity chromatography separates by binding specificity to a ligand specific for the protein of interest. The purified protein is characterized
by ion-exchange chromatography to measure the amount of the protein of interest and the
contaminants.
A nucleotide consists of a nitrogenous base, a pentose sugar, and one or more
phosphate groups. The nucleic acids RNA and DNA are polymers of nucleotides. The
genetic code can be determined by sequencing the four nucleotides that form DNA: A =
Adenine, C = Cytosine, G = Guanine, and T = Thymine. A dideoxynucleoside phosphate
(ddNTP) analog specific for one of the nucleotides interrupts DNA synthesis to prematurely terminate the fragment at that nucleotide, for example A. Different analogs terminate C, G, and T fragments. When each of these radiolabeled (with a radioactive compound) four fragments is separated electrophoretically, it yields the autoradiogram (by
darkening photographic film) pattern in Figure 3.22.
A
G
T
G
T
C
A

Figure 3.22 Nucleotide fragments ending in A, C, G, and T are injected into lanes at the top of electrophoresis columns. The sequence is read from the rows of bands from the bottom up as ACTGTG.

DNA sequencing is automated by labeling each of the four fragments with a

different colored fluorescent tag. Then all four fragments can be combined and analyzed
by a single lane during electrophoresis. The eluting peaks from successively longer fragments are scanned by a 40 mW UV laser beam, which causes them to fluoresce with one
of the four colors. A computer displays the colored peaks and also the sequence. Electrophoresis is carried out in 64 lanes of 200 m thick acrylamide denatured gel between 300
200 mm optical grade glass for 7 h at 1700 V and 3 mA.

3.17 References
Antolasic, F. 1996. What is Mass Spectrometry? [Online]
http://minyos.its.rmit.edu.au/~rcmfa/mstheory.html
Berger, S. A. 1996. Introduction to Bioengineering. New York: Oxford University Press.
Blum, L. J. 1991. Biosensor Principles and Applications. New York: Marcel Dekker.

3-38 Chapter 3 Analysis of Molecules in Clinical Medicine

Bowers, L. D., Ullman, M. D. and Burtis, C. A. 1994. Chromatography. In C. A. Burtis
and E. R. Ashwood (eds.), Tietz Textbook of Clinical Chemistry. 2nd ed. Philadephia:
W. B. Saunders.
Burtis, C. A. and Ashwood, E. R. (eds.) 1994. Tietz Textbook of Clinical Chemistry. 2nd
ed. Philadelphia: W. B. Saunders.
Canh, T. M. 1993. Biosensors. London: Chapman & Hall.
Carrier, R. and Bordonaro, J. 1994. Chromatography. [Online]
http://www.rpi.edu/dept/chem-eng/Biotech-Environ/CHROMO/chromequip.html
Considine, D. M., and Considine, G. D. (eds.) 1995. Van Nostrands Scientific Encyclopedia. New York: Van Nostrand Reinhold.
de la Pea, A., Liu, P., and Derendorf, H. 2000. Microdialysis in peripheral tissues. Advanced Drug Delivery Reviews 45: 189216.
DM Scientific 2000. DS-34 Camera for Electrophoresis Gel Recording by Polaroid.
[Online] http://www.dmscientific.com/ds34.html
Hall, E. A. H. 1991. Biosensors. Englewood Cliffs, NJ: Prentice-Hall.
Hicks, M. R., Haven, M. C., Schenken, J. R., and McWhorter, C. A. 1987. Laboratory
Instrumentation. 3rd ed. Philadelphia: J. B. Lippincott.
Haven, M. C., Schenken, J. R., Tetrayktm J. R. 1995. Laboratory Instrumentation. 4th ed.
New York: Van Nostrand Reinhold.
IQ Scientific Instruments. 2001. The ultimate source for non-glass ISFET technology pH
systems [Online] http://www.phmeters.com
Kurnik, R. T., Berner, B., Tamada, J., and Potts, R. O. 1998. Design and simulation of a
reverse iontophoretic glucose monitoring device. J. Electrochem. Soc. 145: 4119
4125. [Online] http://www.cygn.com/monitor/journals.html
Lehninger, A. L., Nelson, D. L., and Cox, M. M. 1993. Principles of Biochemistry. 2nd
ed. New York: Worth Publishers.
Mallinckrodt. 2001. OxiFirst Fetal Oxygen Saturation Monitoring System [Online]
http://www.mallinckrodt.com/respiratory/resp/.
Melcher, U. 1998. Gel Electrophoresis. Molecular Genetics. [Online]
http://opbs.okstate.edu/~melcher/MG/MGW4/MG422.html
Narayanan, S. 1989. Principles and Applications of Laboratory Instrumentation. Chicago:
American Society of Clinical Pathologists.
Pagana, K. D., and Pagana, T. J. 1995. Mosbys Diagnostic and Laboratory Test Reference. 2nd ed. St. Louis: Mosby.
Peura, R. A. 1998. Chemical biosensors. In J. G. Webster (ed.) Medical Instrumentation:
Application and Design. 3rd ed. New York: John Wiley & Sons.
Primiano, F. P. Jr. 1998. Measurements of the respiratory system. In J. G. Webster (ed.)
Medical Instrumentation: Application and Design. 3rd ed. New York: John Wiley &
Sons.
Taylor, R. F., and Schultz, J. S. 1996. Handbook of Chemical and Biological Sensors.
Bristol UK: IOP Publishing.
Tissue, B. M. 2000. High Performance Liquid Chromatography. [Online]
http://www.chem.vt.edu/chem-ed/scidex.html
Tissue, B. M. 2000. Fluorimetry. [Online] http://www.chem.vt.edu/chem-ed/scidex.html
Tissue, B. M. 2000. Infrared Absorption Spectroscopy (IR). [Online]
http://www.chem.vt.edu/chem-ed/scidex.html

3.17 References

3-39

Tissue, B. M. 2000. Mass Spectroscopy (MS). [Online] http://www.chem.vt.edu/chemed/scidex.html

Webster, J. G. (ed.) 1997. Design of Pulse Oximeters. Bristol, UK: IOP Publishing.
Webster, J. G. (ed.) 1988. Encyclopedia of Medical Devices and Instrumentation. New
York: John Wiley & Sons.
Wheeler, L. A. 1998. Clinical laboratory instrumentation. In J. G. Webster (ed.) Medical
Instrumentation: Application and Design. 3rd ed. New York: John Wiley & Sons.

3.18 Problems
3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9
3.10
3.11
3.12
3.13
3.14
3.15
3.16
3.17
3.18
3.19
3.20
3.21
3.22

Give the equation for Beers law, define each term, and give units.
Explain the operation of a spectrophotometer and its purpose. List the components
of a spectrophotometer.
A sample concentration of 10 mg/dL yields a spectrophotometer transmission of
35%. Assume Beers law holds and calculate the unknown concentration for a
transmission of 70%.
A sample of concentration 20 mg/dL has an absorbance of 0.4 in a spectrophotometer. The sample is then diluted and yields an absorbance of 0.25. Calculate the
new concentration.
Define oxygen saturation and state the physiological meaning of SaO2.
Search the literature for a plot of SaO2 versus PO2 and sketch it.
Describe how NADH is used to measure lactate concentration and why lactate
concentration isnt ascertained by measuring lactate directly.
Describe why creatinine is measured and the technique used to measure it.
Describe why and how to measure urea and the body fluids it can be measured in.
Describe why and how to measure glucose from a drop of blood.
Describe how to measure glucose in automated equipment.
Describe amperometry as used in the PO2 electrode.
Describe the most common enzymatic electrode method for measuring glucose.
Calculate the pH for a hydrogen ion concentration of 107 mol/L.
Draw a pH electrode and explain its principle of operation. Explain why its amplifier input impedance is important. Explain the relation of the CO2 electrode to the
pH electrode.
Explain the principle of operation and give an example of use for flame photometry.
Explain the principle of operation and give an example of use for mass spectrometry.
Explain why and how CO2 is measured by infrared transmission spectroscopy.
Explain why and how N2 is measured by emission spectroscopy.
Explain why and how fluorometry is used. Describe one of the advantages of
fluorometry.
Explain why and how chromatography is used. Explain the two principles that are
the primary factors affecting interactions in chromatography.
Explain how the glucose sensor minimzes sensitivity to PO2 variations.

3-40 Chapter 3 Analysis of Molecules in Clinical Medicine

3.23 Explain why and how electrophoresis is used. List the factors that cause differential
migration rates among the component molecules of a mixture.
3.24 Explain how protein purity is measured.
3.25 Explain how the DNA code is determined.

4
Surface Characterization in Biomaterials and Tissue
Engineering
Jorge E. Monzon

This chapter discusses measurements of molecular variables that are needed for research
and development in the rapidly growing areas of biomaterials and tissue engineering.

4.1 Molecules and Biomaterials

The following sections present some biomaterials fundamentals. They are needed to
understand the importance of measuring certain molecular properties.
4.1.1 Integrated approach
Over the years, a number of different types of materials have been used in medical
practice, either for treating wounds and diseases or for cosmetic purposes.
A biomaterial is a synthetic material used to replace part of a living system or to function
in intimate contact with living tissue (Park and Lakes, 1992). This definition suggests that
the development of materials for biological use can be accomplished only through an
integrated approach of several disciplines: science and engineering, especially material
sciences for detailed study and testing of the structural properties and phenomena of
materials; biology and physiology, to provide the necessary tools for experimentation
based on immunology, anatomy, cellular and molecular biology; clinical sciences, as
many medical specialties are directly related to the actual use of biomaterials (Park,
1995).
A device fabricated with a biomaterial that is placed inside the living system is called an
implant.
4.1.2 Types of biomaterials

Surface Characterization in Biomaterials and Tissue Engineering

Materials for biological use can be classified according to their base structure as ceramics,
composites, metals, and polymers. Table 4.1 summarizes different types of biomaterials
and their most common applications (Silver, 1994; Park and Lakes, 1992). This listing is
not comprehensive, as the search for new materials and their clinical applications is a
dynamic process. Synthetic polymers constitute the vast majority of biomaterials used in
humans (Marchant and Wang, 1994).
Table 4.1 Classification of biomaterials in terms of their base structure and some of their most
common applications.

Biomaterials

Applications

Ceramics
Aluminum oxide
Carbon
Hydroxyapatite

Dental and orthopedic

Composites
Carbon-carbon fibers
and matrices

Heart valves and joint implants

Metals
Atoms
Aluminum
Chrome
Cobalt
Gold
Iridium
Iron
Manganese
Molybdenum
Nickel
Niobium
Palladium
Platinum
Tantalum
Titanium
Tungsten
Vanadium
Zirconium
Metallic alloys
wide variety using metallic atoms

Joint replacement components,

fracture fixation, dental implants,
pacemakers, suture wires,
implantable electrodes

4.1 Molecules and Biomaterials

Polymers
Nylon
Synthetic rubber
Crystalline polymers

Replacement of soft tissues: skin,

blood vessels, cartilage, ocular lens,
sutures
Orthopedic

Figure 4.1 These titanium-alloy joint replacements are an example of the many applications for
metal biomaterials for implantations. (from
http://www.spirebiomedical.com/Biomedical/ionguard.html)

4.1.3 Polymers
Structure
A polymer is characterized by a repeating subunit (monomer) covalently connected to
form a macromolecule.

CH3
CH
Figure 4.2 Polymers are made up of many monomers. This is the monomer for poly(ethylene), a
common biomaterial used for medical tubing and many other applications.

The average number of repeating units is called the degree of polymerization

(Lee et al., 1995). A larger number of these subunits (i.e. bigger molecule) leads to a

Surface Characterization in Biomaterials and Tissue Engineering

polymer of higher molecular weight, and thus to a more rigid material (Park and Lakes,
1992).
Synthetic polymers are obtained through two basic chemical processes: addition
polymerization and condensation polymerization. Some addition polymers are
poly(ethylene), poly(methyl methacrylate), poly(vinyl chloride), and poly(ethylene
terephtalate). Condensation polymers include poly(esters), poly(amides), and
poly(urethanes) (Marchant and Wang, 1994; Park and Lakes, 1992; Silver, 1994).
Complex chemical processes beyond the scope of this book allowed the
development of polymers of very different physical and chemical properties to find
assorted applications in health care. Table 4.2 illustrates the most common polymers and
their clinical uses.
Table 4.2 The clinical uses of some of the most common biomedical polymers relate to their
chemical structure and physical properties.

Biomedical polymer
Poly(ethylene) (PE)
Low density (LDPE)
High density (HDPE)
Ultra high molecular weight
(UHMWPE)

Application
Bags, tubing
Nonwoven fabric, catheter
Orthopedic and facial implants

Poly(methyl methacrylate) (PMMA)

Intraocular lens, dentures, bone

cement

Poly(vinyl chloride) (PVC)

Blood bags, catheters, cannulae

Poly(ethylene terephthalate) (PET)

Artificial vascular graft, sutures,

heart valves

Poly(esters)

Bioresorbable sutures, surgical

products, controlled drug release

Poly(amides) (Nylons)

Catheters, sutures

Poly(urethanes) (PU)

Coat implants, film, tubing

Properties
Biomaterials for implants should be nontoxic, noncarcinogenic, nonallergenic, functional
for its lifetime, and biocompatible (Billotte, 2000).

4.1 Molecules and Biomaterials

Polymeric biomaterials exhibit reasonable cost of processing and can be

manufactured with various shapes showing desired mechanical and physical properties.
In addition, they can be sterilized through different methods (Lee et al., 1995).
However, polymers may loose some of their properties with time, a process
called degradation, which is the consequence of chemical, mechanical, photo or thermal
factors.
Biocompatibility
The most important requirement for an implant is its biocompatibility, that is, the
acceptance of an artificial device by the surrounding tissues and by the body as a whole
(Park, 1995). Greisler et al. (1996) suggest that biocompatibility does not mean lack of
biological responseas might be expectedbut rather the presence of a desirable set of
biological responses.
Bulk and surface properties of polymer biomaterials, which are functions of
their molecular structure and organization, determine the interaction with living
organisms (Marchant and Wang, 1994).

Figure 4.3 This artificial heart valve is coated with a biocompatible material that allows the body
to accept the implant. (from
http://www.sjm.com/products/heartvalves/mechanical/mastersseries.shtm)

4.2 Molecules and Tissue Engineering

As is the case with biomaterials, molecular phenomena also play an important role in
tissue engineering research. We next present some basic aspects of this engineering field.
4.2.1 Tissue engineering
Tissue engineering can be defined as the application of engineering and life sciences
toward the design, growth, and maintenance of living tissues (Berthiaume et al., 1994).
This discipline uses living cells and their extracellular products to develop more
biological replacements instead of using only inert implants (Skalak et al., 1988). Tissue
engineering activities can be performed either in vivo (in the body) or in vitro (in
solution), the latter being of more interest to bioengineers.
Bioartificial tissues are constructed in vitro, are composed of biologic and
synthetic substances, and represent a valid alternative to organ transplantation in cases of
impaired natural organ function, vital organ failure, or organ donor unavailability.
Although many bioartificial tissues are still in a developmental stage, it is expected that
in the future bioartificial organs and tissues will be able to regenerate and perform
complex biochemical functions, similar to the natural organs and tissue they will replace,
a behavior that pure artificial implants cannot exhibit (Berthiaume et al., 1994).
4.2.2 Cellular composites
Tissues are multiphase systems of cellular composites. We can identify three main
structural components: cells organized into functional units, the extracellular matrix, and
scaffolding architecture (Wintermantel et al., 1996).
The three-dimensional structures that characterize bioartificial tissues represent
a cell mass that is greater, by several orders of magnitude, than the two-dimensional cell
cultures traditionally developed by biologists (Berthiaume et al., 1994). Tissue
engineering also concentrates on the development of in vitro models that overcome the
absence of an adequate extracellular matrix in conventional cell cultures (Sittinger et al.,
1996).
Table 4.3 shows some current examples of applications and in vitro research in
tissue engineering (Berthiaume et al., 1994; Palsson and Hubbell, 1995; Skalak and Fox,
1988).
Table 4.3 Some examples of current applications in tissue engineering. Not all of the listed
applications are at the same developmental stage.

Biological system

Example of application

Blood

Hematopoietic (production of red

blood cells by) stem cells culture

4.2 Molecules and Tissue Engineering

Cardiovascular

Endothelialized synthetic vascular

grafts (angiogenesis)
Regeneration of the arterial wall
Compliant vascular prostheses

Liver and pancreas

Bioartificial pancreatic islets

Bioartificial liver

Musculoskeletal

Cartilage reconstruction
Bone reconstruction

Neural

Neurotransmitter-secreting cells
(polymer-encapsulated)
Neural circuits and biosensors
Peripheral nerve regeneration

Skin

Bioartificial skin substitutes

Of the structural components of tissues mentioned earlier, the scaffolding

architecture is, for tissue engineers, a synthetic biomaterial. The interaction of cells with
the biomaterial surface is a main topic in tissue engineering.
The characteristics of the synthetic material will directly affect the recruitment,
attachment, proliferation and differentiation of living cells, that is, defining the quality of
the new tissue. The behavior of molecules from the surrounding tissue and on the surface
of the biomaterial determines the conditions of the interface to which cells respond
(Boyan et al., 1996).
After our review of the main aspects of biomaterials and tissue engineering, it
should be clear that measurements at the molecular scale constitute a key activity that
must be routinely performed by biomaterials and tissue engineers.
In the next sections we discuss some of the most important molecular variables,
and we present the necessary tools for quantitatively studying them.

4.3 Surface Analysis

The successful use of a synthetic polymereither as a biomaterial implant or as a
scaffolding architecture for tissue growthrelies on the qualitative and quantitative
knowledge of several properties of such a polymer.
Numerous physical and chemical variables arising at a microscopic scale will
eventually determine the overall quality of the interaction of synthetic polymers and
living organisms. It is not possible to determine which of these variables prevails over the
others, so researchers are continuously designing instruments, procedures and techniques
for better understanding of these variables, and for relating them to the overall behavior
of material implants.

Surface Characterization in Biomaterials and Tissue Engineering

Surface analysis of biomaterials is a set of procedures to evaluate the

characteristics and properties of biomaterials at a microscopic level.
A complete description of the chemistry and morphology of a surface cannot be
accomplished using one single technique. Surface characterization is based upon partial
descriptions achieved from different methods (Castner and Ratner, 1988).
Basic characterization using microscopy, the study of the chemical composition
by means of spectroscopy, and the evaluation of the polymerwater interface using
contact angle methods, are the main variables and procedures currently used in polymer
surface analysis (Andrade, 1985a).
This section reviews the most common variables of interest and some widely
used instruments to measure them. References describe the principles governing the
procedures and instruments that are presented.
Due to its importance for biomaterials and tissue engineering, we discuss protein
adsorptiona mechanism related to surface behaviorin a separate section.
4.3.1 Topography
Andrade (1985c) points out that the very first characterization to be performed on any
surface is to look at it. Some properties such as color or reflectivity can be observed with
the naked eye. Microscopes are needed for observing other qualities such as roughness,
surface texture, or topography (or microtopography for a molecular scale).
Major developments in topographic studies of synthetic and biologic materials
were achieved after the invention of the electron microscope. Next we discuss the
different types of electron microscopes that are used for observation of surface
topographies.
Transmission Electron Microscope (TEM)
This is the classical electron microscope. Developed several decades ago, TEM made
possible a relevant increase in magnification compared to visible light optical
microscopes. Instead of using a beam of light as the image-forming radiation, TEM relies
on an electron beam that passes through the sample in a vacuum medium to develop a
transmitted image of the specimen.
Resolution and magnification: The resolving power, or resolution, of a microscope is its
ability to discriminate two closely placed structures. Magnification is a measure of the
increase in the diameter of a final image compared to the original. Higher magnification
does not necessarily mean better resolution. Useful magnification is calculated as the
ratio of the resolution of the human eye to the resolution of the lens systems (Bozzola and
Russell, 1991).
The resolving power of a microscope is a function of the wavelength, , of the
source of illumination. For an electron, the wavelength is given by de Broglies equation:

4.3 Surface Analysis

h
mv

9
(4.1)

where h is Plancks constant, m is the mass of the electron, and v is the velocity. Better
resolution is achieved by using lower wavelengths. It is therefore possible to increase the
resolving power of the microscope by increasing the electron velocity and thus lowering
. Electrons accelerated in an electric field will reach a velocity:

v=

2eV
m

(4.2)

where V is the accelerating voltage, e is the charge of the electron, and m is its mass. By
substitution in Eq. (4.1), and by replacing the constants by their values (see Appendix),
we obtain

1.22
V

nm

(4.3)

which shows that short wavelengths can be obtained by using high voltages (on the order
of kilovolts). Equation (4.3) is not useful for calculating the resolution because lens
aberrations, refractive index of the medium, and aperture angles limit the resolution.
However, it is possible to demonstrate that the highest practical resolution of TEM is 0.4
to 1 nm (4 to 10 ), with a practical magnification of 100,000 to 200,000.
Operation: Figure 4.4(a) shows the basic components of a TEM. Not shown is the overall
vacuum system for proper operation. The simplest electron source is a heated, pointed
tungsten wire, although higher resolution microscopes are equipped with a lanthanum
hexaboride (LaB6) cathode to which typically voltages of 40 to 120 kV are applied
(Dykstra, 1992).

10

Surface Characterization in Biomaterials and Tissue Engineering

Figure 4.4 (a) TEM microscope. The electron beam passes through the sample, generating on the
fluorescent screen a projected image of the sample, which can be recorded by photographic means.
(b) SEM microscope. Condenser lenses focus the electron beam on the specimen surface leading to
secondary electron emission that is captured by the detector and visualized on the CRT screen.
Both TEM and SEM operate in a particle free (vacuum) environment.

A set of condenser lenses is used to focus electrons onto the specimen on the
area under examination. Lenses for an electron microscope are not made of glass as for
optical microscopes, but rather they are magnetic coils (solenoids) that can bend the
electron path (Bozzola and Russell, 1991).

4.3 Surface Analysis

11

Below the sample plane, the objective lens assembly focuses and magnifies the
specimen image. By changing the aperture of the objective lens it is possible to control
the specimen contrast and to correct astigmatism. Scattered elements of the electron beam
emerging from the sample are also eliminated by the objective lens system (Dykstra,
1992).
To control the magnification of the image being projected, it is necessary to use
a projector lens, which will also focus the beam of electrons for appropriate intensity
upon the fluorescent screen. This screen can be observed through a viewing window (not
shown) or the image can be photographically reproduced by exposing a photographic
plate immediately beneath the fluorescent screen (Packer, 1967).
Specimen preparation: A meticulous preparation of the sample is required for suitable
TEM observation. This process is lengthy and involves several steps, most of which are
chemical processes: fixation, washing, dehydration, infiltration with transitional solvents
and with resins, embedding, and curing. Once the specimen is chemically prepared it
must be cut into extremely thin slices or sections (from 30 to 60 nm). This procedure is
called ultramicrotomy and is performed to allow the beams of electrons to pass through
the sample material (Bozzola and Russell, 1991). Preparation of suitable thin samples for
TEM studies in the area of biomaterials is hard to accomplish due to the difficulty of
performing ultramicrotomy on elastomeric materials, such as some polymers (e.g.
polyurethanes). There are some alternative methods of specimen preparation that
overcome this difficulty. Another problem with polymers is their low contrast. However,
this can be overcome by using a defocused conventional TEM; a defocused beam also
avoids sample damage or destruction (Goodman et al., 1988).
One major disadvantage of TEM is its limitation for a proper three-dimensional
view (Dykstra, 1992). To obtain information on the surface and near surface morphology
using TEM, the viewing of the sample must be done at several angles. Using multiple
views could permit reconstruction of the three-dimensional structure. However, angle tilt
is severely limited by the loss of resolution, especially with thick samples needing large
tilt angles (Goodman et al., 1988).
For tissue engineers, TEM represents a powerful tool for studying the
intrastructural features of the soft tissuebiomaterials interface, particularly surface
related phenomena such as cell adhesion and biomaterial degradation (Sheffield and
Matlaga, 1986).
Scanning Electron Microscope (SEM)

SEMs and TEMs have many features in commonthey both use electron beams to
visualize a sample. However, they differ in the way the beam interacts with the specimen.
The principle of SEM operation makes it very useful for topographic analysis, as we will
see in the next paragraphs.
Principle: SEM is based upon the interaction of an electron beam with a specimen. The
incident beam (primary electrons) displaces orbital electrons from the sample atoms,
giving rise to secondary electron emission (Figure 4.5)

12

Surface Characterization in Biomaterials and Tissue Engineering

Incident beam
(primary electrons)

Secondary
electrons

Backscattered
electrons

Figure 4.5 Principle of SEM operation. An incident beam of primary electrons displaces orbital
electrons from the sample atoms resulting in secondary electron emission, which is detected for
image formation. Some primary electrons pass by the nucleus to become backscattered electrons.

The number of secondary electrons emitted depends on the electron density of

the specimen, which is related to its elemental composition, so suitable detection of this
secondary emission yields information on the atoms of the surface sample. Some primary
electrons may not collide with orbital electrons. Instead they pass by the nucleus to
become backscattered electrons (Dykstra, 1992).
These are not the only emissions arising at the atomic level, as we explain in
section 4.3.2 (see Figure 4.10), but these two are the ones of interest for SEM operations.
Operations: Figure 4.4(b) shows a schematic diagram of a scanning electron microscope.
Notice the similarities with the TEM in Figure 4.4(a). Some components of the SEM are
identical to those of the TEM, such as a vacuum system for operation in a particle-free
environment. Both microscopes use the same electron source, except that in the SEM
accelerating voltages are limited to around 40 kV. Magnetic lenses of various types are in
the TEM and in the SEM they are all condensers. The purpose of the condenser lenses in
the SEM is to demagnify the electron beam from approximately 50 m at the electron
gun down to about 2 nm, or less, near the specimen, as small spot sizes are essential for
high resolution (Bozzola and Russell, 1991).
Deflector coils bend the primary beam for scanning across the specimen. This
high-energy beam produces secondary electrons near the surface of the sample. Some of
these secondary electrons contact a positively charged detector, also called a collector,
connected to a cathode ray tube (CRT), giving a point of illumination on the CRT screen

4.3 Surface Analysis

13

whose intensity is proportional to the number of secondary electrons from the

corresponding point on the specimen. Deflector coils in the SEM are connected to the
deflector plates of the CRT so that as the SEM beam is scanned over the specimen, the
CRT gun is synchronously swept over the screen. This imaging mode is known as SEI
(secondary electron imaging) (Dykstra, 1992).
Resolution and images: Image resolution in SEM is a more complex concept than in
TEM because time is an integral part of the image formation process in scanning
microscopes (Dykstra, 1992).
Synchronization of beam and CRT plate deflectors account for simultaneous
emission of secondary electrons and their imaging on the CRT screen.
However, some backscattered electrons, as they leave the atom (Figure 4.5) and
just prior to their exit, may interact with electrons some distance from the initial point of
entry in the specimen, which will produce secondary electron emission. These extraneous
secondary electrons produced by backscattered radiation are also detected by the
collector and displayed on the CRT screen, giving a blurred or noisy signal. One way of
reducing the noise is by increasing the true signal-to-noise ratio. Some backscattered
electrons can also be collected by a special detector. By carefully adjusting the condenser
lenses it is possible to obtain high resolution with suitable depth of field for working
distances (distance from the final condenser to the specimen) of approximately 10 nm
(Bozzola and Russell, 1991).
A very attractive feature of the SEM is its ability to generate images that appear
to have three dimensions. The positioning of the primary scanning beam (incident angle)
and the location of the detector will affect the image contrast, and therefore giving a
sense of three dimensions. Secondary electrons are emitted from various parts of the
specimen, but those that face the primary beam and are in line of sight with the detector
will appear bright on the CRT screen, whereas a dark appearance corresponds to areas
not reached by the beam, and not emitting secondary electrons. Medium brightness
represents partial specimen exposure to the beam, with detection of fewer secondary
electrons. This variation in contrast contributes to the generation of an image that appears
to have depth, although that image is visualized on a two-dimensional CRT screen
(Bozzola and Russell, 1991).
Sample preparation: As for the TEM, specimen preparation for SEM viewing is also a
lengthy process that must be carefully carried out (Dykstra, 1992).
The three dimensional effect on SEM-generated images is very useful for
topographical studies. Usually low voltage SEM (~5 kV) provides similar surface
sensitivity to that obtained with XPS (see section 4.3.2). The sample can be probed to
depths of tenths of micrometers making it possible to convert a chemical structure into
topographical relief. In this way, SEM may be used to view a topography that is a
function of the structure with very high surface resolution (Goodman et al., 1988).
Samples must be stable in the high vacuum environment. Mostly dry samples
are used. Additionally, specimens should be conductive (noncharging), therefore a thin
metal coating on the sample is needed (Andrade, 1985c). The remarkable depth of focus
allows the examination of uneven surfaces with the SEM. The analysis of elemental

14

Surface Characterization in Biomaterials and Tissue Engineering

composition of samples can be achieved by using additional detectors, especially for

backscattered radiation (Dociu and Bilge, 1986).
Scanning Tunneling Microscope (STM)

The scanning tunneling microscope, along with the scanning force microscope (SFM)
that we discuss later, belongs to the more general group of scanning probe microscopes
(SPM or SXM).
Very small sensors (probes) scan the surface of the sample in close proximity. In
this way, it is possible to obtain information on surface topography, and on some
mechanical and electronic properties at the atomic level (Marti, 1993).
The STM was initially designed to give a clearer view of the atomic nature of
the surface of solids. Properties of the surface derive from atoms not being completely
surrounded by other atoms, as they are in the interior of a solid (Dykstra, 1992).
Principle of operation: STM operation is based upon the electron tunneling effect, which
is governed by quantum mechanics. This rather complicated function of kinetic and
potential energieswhich is related to the probability of electrons crossing a gap
between two conducting surfacescan be synthesized in our discussion of the STM by
saying that if a voltage is applied between the probe tip and the sample surface, a current
will develop across the junction (tunneling current).
This is due to the fact that a cloud of electrons exists above the specimen surface
and the number of electrons decreases exponentially with the distance from the surface.
Distances of an atomic diameter can cause a relevant decrease in the tunneling current,
thus allowing precise measurements of the vertical position of atoms at the specimen
surface (Dykstra, 1992). Images are formed by monitoring voltages, tunneling currents
and/or probe positions.
The resolution of the STM is a function of the geometry of the tip, the
topography of the sample, and of their electronic structures. For large objects, in the
micrometer range, the geometry of the tip plays a leading role. However, on an atomic
scale, resolution is governed by the stability of the tunnel barrier width (i.e. the distance
between tip and sample). Lateral resolution of the STM (as the tip scans the specimen
surface) is related to the diameter of the tunneling current filament (Marti, 1993).
Calculation of STM resolution is not a straightforward process. The gap between
probe and specimen is typically 1 nm. Stable and precise STMs can measure distance
variations of 1 pm, provided the instrument is in a vibration-free environment (Dykstra,
1992).

4.3 Surface Analysis

15

(a)

(b)
Figure 4.6 (a) An STM probe tip made of tungsten magnified 4,000 times. The tip is very small,
and can be ruined on a sample, which is seen in Figure 4.6(b). (from
http://www.orc.soton.ac.uk/~wsb/stm/photos.htm)

STM instrument: Figure 4.8 shows a block diagram of the STM. Piezotranslators are used
to move the sample relative to the tip rather than vice versa. The piezoelectric effect is
the mechanical expansion and contraction of a material in response to an electric field. In
this way electric voltages are used to generate a three dimensional (3-D) movement of the
piezoscanner holding the probe. A widely used piezoscanner made of ceramic material is
the piezotube, which has inner and outer electrodes for movements along the three
Cartesian axes x, y, and z. The distance between the tip and the sample is kept at 1 nm.
The sample is connected to a tunnel voltage source. The tunneling current from the
sample to the tip is fed into a current-to-voltage converter. The output of the I/V converter
is further processed at the voltage processor block for display. This voltage is also fed
back to the system via the scanner voltage block, which provides the necessary voltages
for movements along the z axis. A scan controller generates the voltages needed for
movements along the xy plane. Visualization and interpretation of the data provided by

16

Surface Characterization in Biomaterials and Tissue Engineering

the STM must be accomplished using image processing methods and algorithms (Marti,
1993).
A
Sliding
surfaces

B
The assembly A slides
inside the assembly B,
driven by the Zapproach drive piezo.

Z-approach
drive piezo
Scanner
piezo
Sample
holder cage

Figure 4.7 This is a sample of a piezotube. There are different approaches, but all use the same
method of two opposing piezoelectric materials to move the sample in each axis. (from
http://www.topac.com/cryosxm.html)

I/V converter

Tip
Sample

Tunnel
voltage source

Voltage
processor

Piezotube

Scanner voltage

Scan controller

Display

Figure 4.8 STM schematics. The tip of a probe scans the surface of the sample. Three-dimensional
movements of the sample under the tip are accomplished using a voltage-controlled piezoscanner.
The tunneling current crossing from the sample to the tip is further processed leading to a
topographical image.

4.3 Surface Analysis

17

Modes of operation: Constant current and constant height are the two modes of STM
operation. In the constant current mode, very high-gain voltage amplifiers are used, so as
to keep the tunneling current constant and to force the tip to follow the contours. By
recording the input of the scanner voltage block (in fact the z voltage amplifier) as a
function of x and y it is possible to determine the topography of the specimen surface.
Any suitable recording device can be used for such purpose: chart recorder, oscilloscope,
or computer. The constant height mode of operation is appropriate only for flat surfaces.
In this mode the tip is held at a constant height and no longer follows the sample
topography. The tunneling current is then recorded as a function of x and y (Marti, 1993).
Three disadvantages of the STM are the need for an environment free of
vibrations, very stable temperatures and the requirement for highly conductive samples to
develop tunneling currents.
The low cost and simplicity of the STM are attractive. However, careful data
acquisition and interpretation are highly recommended (Ratner, 1988).
Scanning Force Microscope (SFM)

The scanning force microscope (SFM), also known as an Atomic Force Microscope
(AFM) was developed after the STM. They share many of the same features. Other types
of force microscopes are the magnetic force microscope, the attractive mode force
microscope, the friction force microscope, and the electronic force microscope. All of
these microscopes measure forces generated between a probe and a sample surface
(Dykstra, 1992).
Theory of the SFM: As with the STM, operation of the SFM is governed by principles of
quantum mechanics. The force between a tip and the sample is used to obtain an image of
the surface topography. In the absence of other magnetic or electrostatic potential, this
small forceon the order of 100 nNdepends upon the interaction potentials between
the atoms of the tip and of the sample (Marti, 1993).
Contact SFMs require the tip to actually touch the sample. These are the most
common microscopes, although noncontact SFMs are also available (Bustamante et al.,
1996).
Instrument operation: Figure 4.9 shows the SFM. A sharp probe tip is mounted on a
flexible cantilever. As the tip moves toward the sample, two types of forces develop: at
large distances the interaction is attractive due to the van der Waals forces; at short
distances repulsive forces develop due to the exclusion principle of quantum mechanics
(Marti, 1993).

18

Surface Characterization in Biomaterials and Tissue Engineering

Laser

Two-segment
photodetector

Lens

Mirror

Cantilever

Sample

Piezoscanner

Figure 4.9 Sketch of an SFM. A laser beam is focused on the cantilever, and reflected back to a
two-segment photodetector. The difference in output from each segment is proportional to the
deflection amplitude of the cantilever scanning the sample.

The attractive and repulsive forces between tip and sample lead to downward
and upward deflection of the cantilever, respectively. Although the tip radius is very
small (~10 nm), the forces acting on it can deflect the entire cantilever. As the sample
scans under the probe, the cantilever is deflected in correspondence with the surface
topography (Bustamante et al., 1996).
Several methods are used to detect cantilever deflections of less than 1 nm. The
most common method is the optical lever, shown in Figure 4.9. A laser beam is focused
on the cantilever and reflected back to a photodiode in the middle of two subdiodes (twosegment photodetector). Any deflection of the cantilever will result in an unbalance in the
number of photons hitting each of the subdiodes, which in turn will unbalance the electric
currents in the photodiodes. This differential current signal at the output of the
photodetector is then proportional to the cantilever deflection and thus to the surface
topography. The differential current signal is amplified for further processing and also
used in a feedback loop (not shown in Figure 4.9) to control the movements of the sample
with respect to the tip according to the modes of operation discussed next. The optical
lever acts as a motion amplifier, as a 100 pm deflection of the cantilever can easily be
detected by the differential photodiode (Bustamante et al., 1996; Marti, 1993).

4.3 Surface Analysis

19

Modes of operation: Similar to the STM, piezoelectric elements are used to control the
motion of the sample or the tip.
In the constant force mode, the contact force between probe and sample is kept
constant by means of an electronic feedback loop that controls the piezoscanner. We said
previously that these forces are in the nanonewton range, but due to the action on tiny
areas of contact (about 0.1 nm2), very high pressures develop. As the tip scans the surface,
an upward deflection of the cantilever indicates a topographic obstacle and the sample
retracts, by means of the feedback loop, to keep the force constant. The movement of the
sample correlates to the surface topography. In the constant height mode the cantilever
can deflect in both directions, and the sample is kept at a constant height. In this mode,
direct recording of the cantilever deflection is performed. A combined mode, the error
signal mode, requires fast feedback response to follow the surface as in the constant force
mode, but recordings correspond to cantilever deflection and not to the sample motion.
This mode allows enhancement of sharp features (Bustamante et al., 1996).
Thermal drift is not usually a problem using optical detectors, although careful
calibration is required to focus and position the laser beam on the cantileveran
operation that requires the help of optical microscopes. In contact SFM (repulsive force
operation), some lateral forces between the tip and the sample may develop, giving
distorted images and thus reduced resolution. These lateral tensions deform the sample.
The magnitude of the lateral forces is a function of the tip dimension and shape, so
careful design of the tip is necessary. One possible solution to the unwanted shearing
forces is offered by noncontact SFMs. Although very useful for imaging soft samples,
they exhibit some mechanical instability due to their attractive force operation
(Bustamante et al., 1996; Marti, 1993).
The AFM is widely used in surface analyses of biomaterials. It has great
advantages over the STM in that it does not need electrically conducting samples. The
AFM can work very easily in liquid (Descouts, 1995).
4.3.2 Chemical composition

Several properties of a biomaterial can be understood by analyzing the chemical

composition in order to identify elements and their quantity in a sample. Accurate
quantitative information on chemical composition is achieved by means of spectroscopic
methods. We present three of these methods.
X-ray Photoelectron Spectroscopy (XPS)

X-ray Photoelectron Spectroscopy (XPS), also known as Electron Spectroscopy for

Chemical Analysis (ESCA), is a widely used technique for surface analysis. Some major
advantages accounting for its broad acceptance as a key tool include: it is nondestructive,
it provides chemical bonding information, it has ease of data interpretation, andexcept
for hydrogen and heliumit is sensitive to all chemical elements (Andrade, 1985b).

20

Surface Characterization in Biomaterials and Tissue Engineering

Researchers rely on XPSs high surface specificity, which takes its information
from within 10 nm (100 ) of the surface, because it allows the studying of biomaterials
interfaces with living tissue. Concentration and chemical state of every element in the
sample can be obtained using XPS (Paynter, 1988).
XPS is based on the photoelectric effect, shown in Figure 4.10 When an X-ray
photon interacts with an atomic orbital electron, if the photon carries enough energy to
overcome the binding energy of the electron, the electron is liberated from the atom as a
photoelectron. The basic equation for this photoemission process and thus for XPS is
E k = h E b s

(4.4)

where Ek is the electron kinetic energy determined by an analyzer as we will explain

shortly, Eb represents the electron binding energy, h is the incident photon energy ( is
the X-ray frequency and h is Plancks constant) (Andrade, 1985b), and s is a work
function required for photoemissions from solids, it is the extra energy that is required to
transfer the electron from the surface to the vacuum. The value of s is predetermined for
each spectrometer.

X-ray photon

(a)

Photoelectron
(b)

(d)

(c)

Fluorescent radiation

Auger electron

Figure 4.10 When an X-ray photon (a) interacts with an atomic orbital electron of the sample, a
photoelectron (b) is emitted verifying Eq. (4.4). The now unstable atom must relax to the ground
state. The relaxation process can be accomplished by either of two mechanisms: (1) an outer orbital
electron releases energy as fluorescent radiation (c) while occupying the place of the emitted
photoelectron, or (2) the excess energy is used to unbind and emit another outer orbital electron
called an Auger electron (d). These mechanisms operate for different sample depths, yielding the
Auger electron emission characteristic of the outermost surface of the sample.

4.3 Surface Analysis

21

The interaction of a monochromatic X-ray photon beam with the surface of the
sample will result in photoelectrons produced with different kinetic energies according to
their orbital position in the atom (i.e. different bonding energies). An electron analyzer,
also known as an electron spectrometer, can measure the number of these electrons.
Processing the electrons kinetic energies yields a spectrum of photoelectron intensity as
a function of binding energy.
Figure 4.11 shows a typical XPS spectrum. Distinct peaks correspond to specific
atomic orbitals so that the binding energy position of each peak allows elemental
identification. Tabulations of orbital binding energies are readily available and usually
preprogrammed into the data system of the instrument. All of the elements in the periodic
table exhibit binding energies in the approximate range of 0 to 1000 eV (electron volts),
therefore positive identification is possible. Further quantization can be achieved by
measuring the area under the peaks (Andrade, 1985b; Paynter, 1988). Many detected
electrons lose energy in interactions with the sample, leading to lower kinetic energies
and higher binding energies recordings (see Eq. (4.4)) adjacent to every intense peak,
giving the spectrum a stair-step shape (Andrade, 1985b).

Intensity (%)

100

Binding energy (eV)

1000

Figure 4.11 A typical XPS spectrum, showing photoelectron intensity as a function of binding
energy. Each peak may correspond to a distinct element of the periodic table or to different orbital
electrons of the same element. Some peaks may also represent Auger radiation.

Figure 4.12 shows the basic schematic of an XPS instrument. The X-ray source
provides a monochromatic photon beam that hits the surface of the sample, which is
placed on a moving holder for appropriate scanning. The resulting photoelectron

22

Surface Characterization in Biomaterials and Tissue Engineering

radiation is captured by an analyzer. The excitation and emission processes take place in
an ultrahigh vacuum (UHV) chamber (not shown in Figure 4.12) for minimal sample
contamination and for easy passage of the photoelectrons to the analyzer (Paynter, 1988).

X-ray source

Retarding grid

Sample holder
Hemispherical
electrostatic
analyzer

Display

Data
processor

Detector

Slit

Figure 4.12 Basic schematics of an XPS instrument. An X-ray beam strikes the sample surface,
giving photoelectron radiation. These electrons enter the hemispherical analyzer where they are
spatially dispersed due to the effects of the retarding grid and of the electrostatic field of the
concentric hemispheres. Ramping voltages at the retarding grid allow kinetic energy scanning. At
the other end of the analyzer electrons are detected, counted, and a spectrum of photoelectron
intensity versus binding energy is displayed.

The analyzer consists of two concentric and electrostatically charged

hemispheres. An electrically biased grid at the entrance of the analyzer retards the kinetic
energy of the photoelectrons, which are then spatially dispersed as they initiate their
semicircular trajectory between the hemispheres. The voltage across the hemispheres is
kept constant (fixed analyzer transmission or FAT mode), therefore the energy resolution
is also constant. Ramping voltages in the pre-retardation stage allows actual kinetic
energy scanning with a constant resolution mode of typically ~0.1 to 1.0 eV.
At the other end of the analyzer an electron multiplier detector increases by
several orders of magnitude the number of electrons reaching the data processing and
display sections, which usually consist of a phosphorescent screen and a detector device

4.3 Surface Analysis

23

that counts the number of light pulses arising at the screen (Andrade, 1985b; Paynter,
1988).
The use of an X-ray monochromator in XPS eliminates undesirable radiation
coming from the X-ray source (see Figure 4.12). Although monochromated beams
minimize the radiation damage to the sample, they also decrease the intensity of the flux
to the sample (Andrade, 1985b). The detection system must compensate for this reduced
intensity in order to obtain suitable signal levels. This is done by incorporating
multichannel photoelectron detectors (Paynter, 1988).
High cost of the instrument, large area of analysis, high vacuum required and
low speed (several minutes per peak) are major disadvantages of the XPS technique
(Andrade, 1985b). Also the behavior of a biomaterial in a vacuum may differ
considerably from its behavior when in contact with living tissues (Paynter, 1988).
Imaging X-ray Photoelectron Spectroscopy (iXPS)

The lack of spatial resolution has also been an inherent weakness of XPS. A system like
ESCALAB, shown in Figure 4.13, uses the capability of parallel imaging to produce
photoelectron images with spatial resolution better than 5 mm.
In parallel imaging method, photoelectrons are simultaneously collected from
the whole field of view. A combination of lenses before and after the energy analyzer
focus the photoelectrons on a channel plate detector. The hemispherical analyzer permits
electrons with only a narrow band of energy to reach the detector. Neither the X-ray nor
the photoelectrons are scanned to produce an image. Spectroscopy and imaging use the
same analyzer to avoid and risk of misalignment. In ESCALAB, images are formed on a
channel plate detector. When using spectroscopy, an array of channel electron multipliers
is used.
The advantage of iXPS is that it produces images much more quickly than the
conventional methods (for comparable spatial resolution) because the entire field of view
is imaged simultaneously. Using an electron flood source, it is possible to get a real time
(i.e. close to TV rate) physical image of exactly the same position as the images. This
makes setting up experiments much quicker. Moreover, the introduction of a magnetic
objective lens has improved the sensitivity for a given spatial resolution enabling the new
technique of imaging of XPS to be developed.

24

Surface Characterization in Biomaterials and Tissue Engineering

Hemispherical
analyzer

Spectroscopy
detector
Lenses

Imaging
detector

Irises
X rays
Sample

Camera
Zoom
microscope

Figure 4.13 Photograph (from

http://www.thermo.com/eThermo/CDA/Products/Product_Detail/1,1075,15885-158-X-1-1,00.html)
and schematics of an ESCALAB. This iXPS instrument offers the capability of parallel imaging,
which obtains positional information from dispersion characteristics of the hemispherical analyzer
and produces photoelectron images with spatial resolution better than 5 m.

Auger Electron Spectroscopy (AES)

4.3 Surface Analysis

25

AES, based on an electron beam and not an X-ray photon beam, and Auger electron
emission (refer to Figure 4.10) are complementary to XPS for surface analysis, being
especially useful for quantitative analysis of certain elements. The Auger electron can
only be characterized by its absolute kinetic energy and not by a binding energy. AES
found practical applications earlier than XPS, but due to the need of electron
bombardment to the sample (instead of photons), it is a highly destructive technique not
recommended for modern polymer surface analysis (Andrade, 1985b). AES is also
known as SAM (Scanning Auger microprobe) (Ratner, 1988).
Secondary-Ion Mass Spectroscopy (SIMS)

Secondary-ion mass spectroscopy (SIMS) is another technique for studying the nature of
biomaterial surfaces. Due to its destructive nature, it has been applied to biomedical
polymers. However, it can provide useful information for the understanding of the
composition and structure of an implant material surface (Andrade, 1985c; Castner and
Ratner, 1988).
SIMS is based on the principle of mass spectrometry, a proven sensitive and
powerful analytical technique, in which sample molecules are ionized. If enough ionizing
energy is present, sample molecules are even dissociated, forming ionic fragments. The
mass-to-charge ratios of the molecular ions and fragment ions are determined and plotted
against abundance (mass-spectrum). The pattern of product ions formed is dependent on
the structure of the intact compound (Wait, 1993).
Figure 4.14 illustrates the SIMS procedure and instrument schematics. A SIMS
instrument, and in general all mass spectrometers, consists of a means of ion generation,
a mass analyzer for their separation, and an ion detector (Wait, 1993). As with XPS,
ultrahigh vacuum (UHV) chambers are used to avoid not only external contamination but
also collisions with particles that may be present in a nonvacuum environment.

26

Surface Characterization in Biomaterials and Tissue Engineering

Ion
detector

Ion generator

Processing

Mass
analyzer

Display

Secondary
ions

Primary
ions

Sample
Figure 4.14 Schematic diagram of a SIMS instrument. Bombardment of primary ions on the
sample surface leads to secondary ion emission. A mass analyzer separates these ions in terms of
their mass-to-charge ratio. The ion detector converts this ionic current into an electric signal for
further processing. The display presents the SIMS spectra, consisting of the count of ions versus
their mass-to-charge ratio.

Ion generator: The generator or gun provides a beam of charged particles that once
accelerated into the surface sample, lead to the ejection of secondary ions. The size,
energy, mass, and flux of the incident beam are directly proportional to secondary ion
emission and sample damage. A good compromise between suitable secondary emission
and damage is a 2 keV Xe+ ion beam which gives a sampling depth of approximately 1
nm (10 ). The use of such low-energy inert gas ion beams is called static SIMS and is
the most commonly used technique. It allows the analysis of the topmost surface of
insulating materials such as polymers. A high-energy and high-flux beam technique,
called dynamic SIMS, is used to profile the depth of minerals and semiconductors
(Andrade, 1985c).
The use of a charged incident beam as mentioned previously can significantly
alter the electrical characteristics of the sample surface, particularly insulators, as surface
charge buildup will affect the secondary ion detection. One way of avoiding this charge
buildup is by using a neutral atom incident beam, a technique referred to as fast atom
bombardment (FAB) SIMS (Castner and Ratner, 1988).

4.3 Surface Analysis

27

Charge neutralization: For suitable reading of secondary ion emission, the charge that
builds up on the surface of an insulating sample must be neutralized. This can be
accomplished by a variety of devices that generate electron beams for the case of
positively charged incident ion beams. Even for the case of neutral FAB SIMS, the use of
neutralizing electrons has shown to increase the quality of detected data (Castner and
Ratner, 1988)
Mass analyzer: The counts of secondary emitted ions versus their mass-to-charge ratio
(measured in Daltons, where 1 D = 1/12 the mass of C12) constitute the data obtained
through a SIMS instrument.
The well known expression for kinetic energy in terms of mass and velocity can
be applied to an ion, for which its number of charges z (in units of the charge of an
electron) and its accelerating voltage V are known, leading to Eq. (4.5)
zV =

1 2
mv
2

(4.5)

which can be rearranged to give the fundamental mass spectrometer equation

m 2V
= 2
z
v

(4.6)

where v is the ion velocity and m its mass (Wait, 1993).

The quadrupole mass spectrometer, which determines mass divided by charge
m/z by scanning voltages, is used for analyzing fragments of up to 500 D. Most polymers
yield fragments in this range. Quadrupole spectrometers are commonly used because they
are inexpensive and most polymers yield fragment ions in their detection range (Castner
and Ratner, 1988).
For sensitivities higher than 500 D, the time-of-flight (TOF) mass analyzer is
used. The TOF mass analyzer measures velocity instead of voltages. It measures the time
through a path of fixed length (see Eq. 4.6) to calculate the ratio m/z. For precise time
measuring, pulsed ionization techniques must be employed for primary beam generation,
which accounts for the higher cost of TOF-based instruments.
Data processing and interpretation: Once the ions are separated according to their massto-charge ratio m/z, a detector is used to convert the ion current into an electric signal by
a type of electron multiplier. The electric signal is amplified, computer processed, and
visualized.
In modern instruments, computers are used to select parts of spectra for analysis
and also to control some instrument features such as beam generation and intensity,
neutralization energies, and sample scanning (Castner and Ratner, 1988; Wait, 1993).
Static SIMS is used for determining the types, structures, and quantity of the
chemical compounds on a surface. Note that relating the numerous peaks of the SIMS
spectra to the surface chemical group from which they originated is not an easy task. This

28

Surface Characterization in Biomaterials and Tissue Engineering

is mainly due to the different ionic particlessome of them fragments of molecules

that are detected.
Reference materials, with standard or at least well known spectra, must be used
for comparative purposes. Nevertheless, careful SIMS experiments allow full
identification of polymeric chains, and also some composition up to a depth of 1.5 nm
(15 ). SIMS complements XPS for polymer composition studies (Castner and Ratner,
1988).
ISS: Another ion beam technique called Ion Scattering Spectroscopy (ISS) consists of a
low-energy inert gas beam that interacts with the sample surface atoms. Incident ions lose
energy in these elastic collisions, a loss that depends on the mass of the atom in the
surface. There is no secondary ion emission as in SIMS, and therefore primary ions are
the ones being detected in ISS (Andrade, 1985c).
Infrared Spectroscopy (IR)

Next we present some infrared spectroscopy techniques, which are usually combined to
allow surface analysis.
Infrared spectroscopy is a very versatile, fast, and inexpensive method that leads
to conclusive molecular identification of samples in any of the three phases: gas, liquid,
or solid (Grim and Fateley, 1984).
Dispersive infrared spectroscopy: The infrared region of electromagnetic radiation
includes wavelengths in the range of 700 nm to 1.0 mm. Usually only the mid-infrared
region is used for measuring, and this includes wavelengths from 2.5 m to 50 m
(Knutson and Lyman, 1985). For convenience, spectroscopists use the wavenumber,
expressed in cm1 and defined as 1/, rather than wavelengths. The near infrared region
(NIR) is conventionally defined as the region from 20,000 to 4,000 cm1; the far infrared
(FIR) region is defined to be from 200 to 10 cm1 (Grim and Fateley, 1984).
The underlying principle of infrared spectroscopy states that the absorption of
electromagnetic radiation of the appropriate energy excites the atom. This absorbed
energy is converted to vibrational and rotational motion governed by quantum mechanics
rules. The absorption process depends on the relative masses and geometry of the atoms
and on the forces between bonds within the sample molecules. The wavelengths of the
radiation that can excite a molecule from one vibrational level to another fall in the
infrared region. The vibrational and rotational motions take place at particular
frequencies (or wavelengths) according to the molecular structure. The absorbed energy
associated with the vibration is described by quantum mechanics in terms of discrete
vibrational energy levels (Knutson and Lyman, 1985).
A spectrum (energy vs. wavelength) can be obtained in such a way that it allows
identification of the structural features of the unknown sample compound. An unlimited
number of spectral libraries are available to verify tentative identifications (Grim and
Fateley, 1984).

4.3 Surface Analysis

29

The dispersive infrared spectroscope is similar to the spectrophotometer (see

Chapter 3). They both measure absorbance or transmittance as a function of wavelengths
(or wavenumbers), and they use a prism or grating to disperse radiation from the source
into spectral elements. Dispersive spectrometry has some limitations, which are
overcome by the use of interferometric techniques as we discuss next.
Interferometric infrared spectroscopy: The basics of interferometry are related to the
interference of monochromatic radiation: two waves will interfere with each other if their
optical paths intersect. This interference can be constructive (sum of amplitudes of
radiations in phase) or destructive (amplitudes of out of phase radiations cancel each
other). For intermediate phase differences, the resulting amplitude of the interacting
radiation waves will be a function of the phase angle. An interferogram is the plot of
amplitude vs. phase (Grim and Flateley, 1984).
A major component in interferometry instruments is the interferometer, a device
for developing a difference in phase between two waves (Grim and Flateley, 1984).
Figure 4.15 shows schematically the optical configuration of a Michelson interferometer,
the first designed and still the most common interferometer in use today. It consists of a
beamsplitter set at 45 from a set of two mirrors, one fixed, and the other able to slide
perpendicularly to the fixed mirror. The beamsplitter transmits half of the source
radiation to the fixed mirror and the other half to the sliding mirror (Fink and Larson,
1979). A phase difference between the beams can be induced by sliding the mirror. The
changes in the mirror position change the path length for each beam. The beams are
recombined before passing through a sample to reach the detector. If the path lengths to
the fixed and sliding mirrors are the same, constructive interference occurs to yield a
maximum at the detector. If the path lengths differ by /2, destructive interference occurs
to yield a minimum at the detector. For a monochromatic source, as the sliding mirror
moves distance x, the detector yields an output I = I0 cos(2x/). Thus by counting
maxima as the sliding mirror moves, we can measure the number of wavelengths of
movement.

30

Surface Characterization in Biomaterials and Tissue Engineering

Infrared
detector

Sliding
mirror

Beamsplitter

Source

Fixed mirror
Figure 4.15 Michelson interferometer. A beamsplitter transmits half of the source radiation to the
fixed mirror and the other half to the sliding mirror. A phase difference between the beams can be
induced by sliding the mirror, which causes detection of the two beams at different times. The
detector provides the interferogram, a plot of energy as a function of differences in optical paths.
Beams have been slightly shifted in the drawing to allow easy following of their path.

Fourier Transform infrared spectroscopy (FT-IR): For a polychromatic source, the

interferogram will show a center burst (maximum) only when the two path lengths are
equal. We use equal path lengths with no movement. Then if absorption from one path is
present at one wavelength, an inverted cosine wave will appear across the interferogram.
We can use a mathematical procedure: the Fourier Transform (FT) or the Fast Fourier
Transform (FFT) to measure all wavelengths simultaneously and convert from the spatial
domain of the interferogram to the absorption spectrum of spectroscopy (Grim and
Fateley, 1984).
This operator allows the display of the interferogram (and in fact of any time
signal) in the frequency domain. The interferogram represents all wavelengths in one
graph; this is an advantage over dispersive spectroscopy in which energy levels are
measured for each wavelength at a time. Calculations for the FT for displaying
interferograms in the frequency domain were a lengthy and tedious process that has since
been all but eliminated by the use of digital computers. This powerful mathematical tool
has also allowed the use of interferometry in the far infrared region, which required
almost endless calculations. It has improved the benefits of interferometry over dispersive

4.3 Surface Analysis

31

infrared spectroscopy, and most current interferometric instruments have a built-in

computer for FT calculations (Buijs, 1984).
Attenuated internal total reflection (ATR): Total internal reflection is a special case of the
more general phenomenon of reflection and refraction of incident electromagnetic
radiation at an interface of two media having different indexes of refraction (Knutson and
Lyman, 1985).
Figure 4.16 shows that when an incident beam with an angle traveling in a
medium of refractive index c encounters the interface with a second medium of
different refractive index s, the beam will be reflected back into the first medium and
transmitted into the second medium with an angle other than the incident (refraction)
according to Snells Law of Refraction:

c sin = s sin

(4.7)

where the indexes of refraction represent relationships of the speed of light in the medium
to the speed of light in a vacuum. From Eq. (4.7) and from Figure 4.10 it becomes
evident that if the first medium is denser that the second (i.e. c >s), then is greater
than . Also, there is a value for that will make = 90 and is called the critical angle
c. Eq. (4.7) becomes

c = sin 1 s
c

(4.8)

Values of larger than c will cause total reflection of the incident beam (i.e. no
refractive beam develops).

32

Surface Characterization in Biomaterials and Tissue Engineering

Figure 4.16 When an incident beam traveling at an angle in a medium of refractive index c
encounters another medium of refractive index s, it will reflect in a direction given by and
refract in the direction given by , verifying Snells Law of Refraction (Eq. (4.7)).

We have highly simplified the presentation of these phenomena. A more

rigorous treatment would have required the use of tools provided by electromagnetic
wave theory.
Surface infrared spectroscopy, FTIR-ATR: The combination of the techniques that we
have just presented represent a very useful method for studying the surface properties of
materials.
The electromagnetic radiation is totally internally reflected (ATR) through an
optically transparent material in contact with the sample (see Figure 4.16). The
reflectivity exhibited by the surface depends upon the interactions of the electromagnetic
field established within the sample. Using infrared spectroscopy (IR) it is possible to
obtain a spectrum representing the characteristics of the sample surface within a finite
distance of the interface, providing information on its chemical composition (Knutson
and Lyman, 1985).
Finally, the addition of the Fourier Transform led the method known as Fourier
Transform Infrared Spectroscopy-Attenuated Total Internal Reflection, FTIR-ATR,
which ultimately combines the versatility of the infrared spectrum and the convenience of
computers for fast digital processing, represents a key technique for surface analysis.

4.3.3 Polymerwater interface

4.3 Surface Analysis

33

Biomedical polymers are usually implanted within a liquid environment. The polymerliquid interface in terms of surface free energy and surface tension can be accurately
measured with the contact angle method.
Contact angle method

The contact angle method requires relatively simple and inexpensive equipment, although
it is one of the most sensitive methods used to obtain information about the outermost
layer of solid surfaces (Andrade et al., 1985).
This method allows the determination of the excess free energy per unit area, the
surface tension, whose SI unit is N/m. The surface is actually a discontinuous boundary
between three different phases: liquid, gas, and solid. A surface tension develops in these
phase boundaries when the equilibrium is altered giving rise to an excess energy, which
will minimize the surface area. Adsorption (attraction of foreign materials) or
chemisorption (bonding with the adsorbent) are other means of minimizing the surface
energy (Park and Lakes, 1992).
Figure 4.17 illustrates the three surface tension components that interact to limit
the spread of a drop on top of a rigid surface. At equilibrium, Youngs equation is

SV SL = LV cos

(4.9)

where is the interfacial free energy for the solid-vapor, solid-liquid, and liquid-vapor
boundaries, and is the contact angle (Andrade et al., 1985). Rearranging Eq. (4.9) leads
to
cos =

SV SL
LV

(4.10)

LV

SV

SL

Figure 4.17 Surface tension components of a three-phase system to limit the spread of a drop on
top of a surface. is the interfacial free energy for each of the phases. is the contact angle.

34

Surface Characterization in Biomaterials and Tissue Engineering

The values of describe the wetting characteristic of a surface; a complete

wetting is represented by a value of zero for . Nonwetting corresponds to > 90 and to
a more spherical drop (see Figure 4.17). Equation 4.10 shows that the value of
represents ratios and not absolute values of surface tension. For a contact angle greater
than zero, the lowest surface tension of a liquid in contact with a solid surface is called
the critical surface tension, c, which can be obtained by measuring contact angles of
homologous liquids (Park and Lakes, 1992).
Several conditions must be satisfied for a straightforward interpretation of the
contact angle: the solid surface is rigid, very smooth, uniform, homogenous, and does not
interact with the liquid other than the three-phase equilibrium; the liquid surface tension
is well known, constant, and the liquid vapor does not adsorb on the solid surface to
change its surface free energy (Andrade et al., 1985).
The contact angle can be measured, among another methods by: (1) direct
microscopy with a goniometer or protractor for direct measurement of the angle, (2)
calculation of the angle from known dimensional measures of the drop, such as diameter,
volume and trigonometric relationships and (3) capillarity of a liquid of known surface
tension (Andrade et al., 1985).

4.4 Protein Adsorption

Adsorption is the increased concentration of a substance at the surface of a substrate.
The interaction of proteins with solid surfaces, which is protein adsorption,
represents an important process for the fields of biomaterials due to its role in blood
coagulation and in the reaction of tissues to artificial implants. For tissue engineers, cellsurface adhesionwhich is mediated by an interlayer of adsorbed proteinis usually
required for cell culture in vitro (Andrade, 1985d).
Diverse processes are affected by protein adsorption: thrombosis and hemostasis,
healing of hard and soft tissues, infection and inflammation. Some types of artificial heart
valves require permanent anticoagulation treatment, which in turn requires knowledge of
the cellprotein interface. For orthopedic applications, surface properties affect bone
healing (Chinn and Slack, 2000).
It is generally accepted that the first interaction of blood with a foreign surface is
the adsorption of proteins and that the nature of the adsorbed proteins will affect further
interactions of the blood (Choi and Callow, 1994). This blood compatibility or
incompatibility is considered to be related to adsorption processes involving gamma
globulins, fibrinogen, fibronectin, albumin, and several other proteins (Andrade, 1985d).
Tissue engineering promotes new tissue formation. Whether or not the
biomaterial used for such purpose is biodegradable, its surface properties (affected by
adsorbed proteins) will influence the initial cellular events (cell attachment) at the cellmaterial interface (Boyan et al., 1996). Implantation of biomaterials or engineered tissues,
for example, generally leads to an inflammatory response aimed at destroying the foreign
material. A central aspect of inflammatory responses is the adhesion of leukocytes to
blood vessel walls before emigrating into tissues. This cell adhesion to the vessel walls is
mediated by a family of protein receptors (Jones et al., 1996).

4.4 Protein Adsorption

35

Adsorbed proteins may exhibit a complex behavior not only related to the direct
interaction of surface and proteins, but also to the structural re-arrangements (i.e.
conformational changes, see Section below) that proteins may undergo during the
adsorption process (Boyan et al., 1996).
4.4.1 Protein molecule
Proteins are polymers of biologic origin, and among organic molecules they constitute
the largest group.
Carbon, nitrogen, oxygen, and hydrogen atoms are the basic elements in a
protein molecule; in many cases, sulfur atoms are also a part of the protein molecule.
Proteins, as we mentioned, are polymeric chains in which the subunits are called amino
acids. There are 20 different amino acids. Figure 4.18 shows the structure of an amino
acid molecule, which has a central carbon atom. An amino group (NH2), a hydrogen
atom, and an acid group (COOH or carboxyl group) are bonded to the central carbon
atom. Also in Figure 4.18, R represents the rest of the amino acid molecule, a part that is
different for each of the 20 amino acids.

NH2

OH

Figure 4.18 The amino acid molecule. To a central carbon atom, an amino group, a carboxyl group
and a hydrogen atom are bonded. R represents the rest of the molecule, which is different for each
of the 20 amino acids.

Different amino acids combine to form polypeptides and these in turn give rise
to the very high number of complex structures (proteins). This combination of
polypeptides will give to each protein a distinct three-dimensional shape. This spatial
relationship between the amino acid chains giving rise to the structure of a protein is
called conformation. On the surface of a protein molecule there are hydrophobic, charged
and polar domains. The particular regions of a protein to which antibodies or cells can
bind are called epitopes. The existence, arrangement, and availability of different protein
epitopes is called organization (Chinn and Slack, 2000).
4.4.2 Protein adsorption fundamentals

36

Surface Characterization in Biomaterials and Tissue Engineering

For a single protein solution, the rate of adsorption to the substrate depends upon
transport of the protein to the substrate. Four primary transport mechanisms have been
identified: diffusion, thermal convection, flow convection, and combined convectiondiffusion. For constant temperature and static systems, transport is exclusively by
diffusion, and the net rate of adsorption can be described by Langmuirs theory of gas
adsorption (Chinn and Slack, 2000).
rA = k A C b (1 ) k D

(4.11)

where rA is the net rate of adsorption, kA is the adsorption rate constant, Cb is the bulk
concentration of the protein in solution, is the fraction of surface occupied by the
adsorbed molecules and kD is the desorption rate constant. At equilibrium, the rate of
adsorption equals the rate of desorption, so the net rate rA is zero and from Eq. (4.11),
the fractional surface coverage is:

kA

Cb
kD
k
1 + A Cb
kD

(4.12)

Equation (4.12) is called the Langmuir Adsorption Isotherm. It is most applicable to

dilute solutions (Chinn and Slack, 2000).
The initial adsorption and the eventual deposition of various proteins are
different. The Vroman effect designates a set of transient phenomena related to protein
adsorption: the layer composition on the surface appears to be evolutionary and varies
from surface to surface, and that proteins may replace one another in some ordered
sequence during adsorption, giving rise to the concept of protein multilayers. For the
special case of blood compatibility of a material, the sequence of different protein
interactions is not well understood (Choi and Callow, 1994).
4.4.3 Protein adsorption measurements

In order to understand protein adsorption, quantitative evaluation of the adsorption

process must be performed, along with information on the properties and characteristics
of the protein molecule itself that will affect the adsorption mechanisms.
Although a set of useful information is needed from a protein adsorption
experiment, the available techniques are limited and in general do not provide all the
information desired. The amount of adsorbed protein (adsorbed isotherms) can be
measured using XPS and ATR-IR (see section 4.3.2). Adsorption as a function of time is
also measured using ATR-IR and fluorescence spectroscopy. The conformation and
conformational changes (denaturation) of the protein molecule can be determined using
IR techniques (section 4.3.2), circular dichroism (CD) and calorimetry. The thickness of
the adsorbed layer can be measured by ellipsometry. For studying the heterogeneity of

4.4 Protein Adsorption

37

adsorption, electron microscopy (section 4.3.1) and microautoradiography are used. The
use of radiolabeled proteins allows direct measurement of protein adsorbed to a substrate.
Differential Scanning Calorimetry (DSC)

Protein adsorption involves redistribution of charge, changes in the state of hydration and
rearrangements in the protein structure (Norde, 1985). The result of these processes is a
generation of heat, which can be measured in a calorimeter. Most commercial
calorimeters are of the conduction type. The heat developed in the reaction vessel is
completely transferred through a thermopile to a surrounding heat sink of large capacity.
The voltage signal generated by the thermopile is proportional to the heat flux. In a twin
calorimeter, two vessels are within a single heat sink. The reaction takes place in one
vessel and the other vessel serves as the reference. The detectors are in opposition so the
differential signal indicates the heat from the reactor vessel and rejects disturbances from
the surroundings that affect both vessels (Feng and Andrade, 1994).
Circular Dichroism (CD)

Many biologically important molecules are not functionally identical to their mirror
images. They are called optically active or chiral. These asymmetric molecules exhibit
optical rotation. That is, a vertical band of polarized light will rotate to the left or right if
passed through an optically active substance. For left- and right-circularly polarized light
(LCP and RCP), they exhibit circular dichroism (CD). This is the difference in absorption
of LCP and RCP light (Hatano, 1986). Measurements can also be carried out in the
infrared (Ferraro and Basile 1985).
Total Internal Reflection Fluorescence (TIRF)

Most proteins fluoresce naturally due to the intrinsic fluorescence of one or more of the
aromatic amino acids comprising them: tyrosine (Tyr), phenylaline (Phe), or tryptophan
(Trp) (Hlady et al., 1985). Light from xenon, or other light source, passes through an
excitation monochromator to emit a narrow band of usually ultraviolet wavelengths. The
protein in a flow cell fluoresces at a longer wavelength. The emission passes through an
emission monochromator and is measured by a photomultiplier.
Ellipsometry

Ellipsometry operates on the principle that light changes its polarization upon reflection
from a solidsolution interface. It requires optically flat, reflective surfaces and provides
information on the thickness and optical contents of adsorbed film as a function of time.
Morrissey (1977) used ellipsometry to measure differences in adsorbed amount and
thickness of gamma globulin onto flat silica plates.

38

Surface Characterization in Biomaterials and Tissue Engineering

Autoradiography

The distribution of a radiolabeled compound within a specimen can be imaged using

autoradiography, which is more sensitive than pulse counters (Dykstra, 1992).
Radioisotopes are usually administered at 10 to 40 Ci/g (Ci = curie) body weight for in
vivo studies. Silver halide photographic emulsions are placed over the radiolabeled
specimen, then developed. The method is slow and exposure times can be up to 3 weeks.
Radiolabeling

Radiolabeling permits measurement of the affinity for a given protein onto materials
(Baquey et al., 1991). During testing, peristaltic pumps cause biological fluids to flow
through tubing segments of material samples such as PVC, silicone elastomer,
polyurethane or polyethylene. Human albumin may be labeled with 99mTc or radioiodinated. Detectors based on junctions of semiconductors are preferred because of their
good energy resolution. A scintillation camera (used for viewing emissions of radioactive
substances) can be used to image the amount of adsorbed protein on a given surface and
to study its distribution on this surface.

4.5 Molecular Size

For molecular size below a few micrometers, or 10000 D (daltons), particle size can be
measured by light scattering. A laser emits polarized light focused on the solution in a
scattering cell. Light intensity from a particle in solution that is small compared with the
wavelength, , of the incident light is proportional to the concentration, c, multiplied by
the molecular weight Mw. This is known as static or Rayleigh scattering.
In dynamic light scattering, the diffusion constant of the molecules moving in
the solution is calculated from the autocorrelation function of the scattered light. This is
also known as quasi-elastic or photon correlation spectroscopy (PCS).
The thermal agitation of particles causes the random movements of Brownian
motion. Coherent light from two particles scatters and can add or interfere at the
photosensor. Photomultiplier tubes or low-noise, high gain avalanche photodiodes (APDs)
measure the scattered light at various angles. The photosensor output fluctuates as the
particles move relative to each other. Small particles move rapidly whereas large particles
move slowly. Thus the fluctuations are slow for large particles because they do not move
very far between measurements. Photon counter signals are amplified then discriminated
to produce pulses, which are fed to digital signal processor (DSP) correlator. Figure 4.19
shows that the autocorrelation function (ACF) versus time yields an exponential decay
from which particle size is calculated (Brown, 1993).
G() = <I(t) I(t + )>

(4.13)

39

4.5 Molecular Size

where:
G() = ACF
I(t) = intensity at time t
I(t + ) = intensity at (t + )
= delay time
< > = time average

G()

Figure 4.19 The autocorrelation function G() is 1 when two signals have delay time = 0, then
decays to 0 for long delay time.

A PCS measurement can be thought of as a measure of the time it takes for a

particle to diffuse a distance 1/K, where K is the scattering vector
K = 4n

sin ( / 2)

(4.14)

where:
n = refractive index of the diluent (n = 1.333 for water)
= scattering angle
= wavelength ( = 633 nm for HeNe laser)
The ACF function for identical particles is a single decaying exponential
G() = exp(2)

(4.15)

where:
G() = ACF
= decay constant of particle
Note that when G() = 0.37, 1/(2) = . We can then use to calculate the
diffusion coefficient DT = /K2. From the diffusion constant, we can use the Stokes
Einstein equation to calculate the diameter of a spherical particle.

40

Surface Characterization in Biomaterials and Tissue Engineering

DT =

kT
3d

(4.16)

where DT is the diffusion coefficient, the Boltzmann constant is k = 1.38 1023 J/K, T is
the absolute temperature, is the viscosity of the medium ( = 0.001 Pas for water at 20
C), and d is the diameter of the particle.
Example 4.1 Given autocorrelation function data obtained from experiment, calculate the
particle size of polystyrene latex in the sample solution. The data were obtained at a
temperature of 20 C with = 90 and = 632.8 nm. Table 4.4 gives the data for and
the corresponding intensity. The solution is considered water (n =1.33, = 0.001 Pas).
Table 4.4 Experimental data for Example 4.1.

0
0.001
0.002
0.003
0.004

G()
1
0.35
0.15
0.051
0.012

The data from Table 4.4 yield an exponential plot, but it doesnt provide a useful
mathematical model for calculation. Therefore, take a natural log of the data and then
linearize the data using mathematical software or, for example, LINEST in Microsoft
Excel to obtain a linear equation. The given data yield the linear approximation
ln (G ( )) = 1077.2( ) + 0.0852
To calculate the diffusion coefficient, needs to be calculated first. Since G() = 0.37
when 1/(2) = , the linear approximation can be used to find when G() = 0.37.

ln (0.37) 0.0852
= 1.00209 * 10 3
1077.2

Now using the relationship 1/(2) = , = 498.956.

Solve for K using Eq. (4.14) and the known constants
K = 4n

sin ( / 2)
sin (45)
= 4 (1.33)
= 1.87 *10 7
9

632.8 *10

Finally obtain the diffusion coefficient DT = /K2 =1.4271012 and solve for the particle
diameter using the rearranged StokesEinstein equation, Eq. (4.16)

4.5 Molecular Size

d=

41

1.38 10 23 (293)
kT
=
= 301 nm
3DT 3 (0.001) 1.427 10 12

4.6 References
Andrade, J. D. 1985a. Introduction to surface chemistry and physics of polymers. In J. D.
Andrade (ed.) Surface and Interfacial Aspects of Biomedical Polymers, Vol. 1
Surface Chemistry and Physics. New York: Plenum.
Andrade, J. D. 1985b. X-ray photoelectron spectroscopy (XPS). In J. D. Andrade (ed.)
Surface and Interfacial Aspects of Biomedical Polymers, Vol. 1 Surface Chemistry
and Physics. New York: Plenum.
Andrade, J. D. 1985c. Polymer surface analysis: conclusions and expectations. In J. D.
Andrade (ed.) Surface and Interfacial Aspects of Biomedical Polymers, Vol. 1
Surface Chemistry and Physics. New York: Plenum.
Andrade, J. D. 1985d. Principles of protein adsorption. In J. D. Andrade (ed.) Surface and
Interfacial Aspects of Biomedical Polymers, Vol. 2 Protein adsorption. New York:
Plenum.
Andrade, J. D., Smith, L. M. and Gregonis, D. E. 1985. The contact angle and interface
energetics. In J. D. Andrade (ed.) Surface and Interfacial Aspects of Biomedical
Polymers, Vol. 1 Surface Chemistry and Physics. New York: Plenum.
Billotte, W. G. 2000. Ceramic biomaterials. In J. D. Bronzino (ed.) The Biomedical
Engineering Handbook. 2nd ed. Boca Raton, FL: CRC Press.
Baquey, C., Lespinasse, F., Caix, J., Baquey, A. and Basse-Chathalinat, B. 1991. In Y. F.
Missirlis and W. Lemm (eds.) Modern Aspects of Protein Adsorption on
Biomaterials. Dordrecht, Holland: Kluwer Academic Publishers.
Berthiaume, Fl, Toner, M., Tompkins, R. G. and Yarmush, M. L. 1994. Tissue
engineering. In R. S. Greco (ed.) Implantation Biology: The Host Response and
Biomedical Devices. Boca Raton, FL: CRC Press.
Boyan, B. D., Hummert, T. W., Dean, D. D. and Schwartz, Z. 1996. Role of material
surfaces in regulating bone and cartilage cell response. Biomaterials, 17 (2): 13746.
Bozzola, J. J. and Russell, L. D. 1991. Electron Microscopy: Principles and Techniques
for Biologist. Boston: Jones and Bartlett Publishers.
Brown, W. (ed.) 1993. Dynamic Light Scattering: The Method and Some Applications.
Oxford: Clarendon Press.
Buijs, H. 1984. Advances in instrumentation. In T. Theophanides (ed.) Fourier
Transform Infrared Spectroscopy. Dordrecht, Holland: D. Reidel Publishing.
Bustamance, C., Eire, D. A. and Yang, G. 1996. Scanning force microscopy of biological
macromolecules: present and future. In H. C. Hoch, L. W. Jelinski and H. G.
Craighead (eds.) Nanofabrication and Biosystems: Integrating Materials Science,
Engineering and Biology. New York: Cambridge University Press.
Castner, D. G. and Ratner, B. D. 1988. Static secondary ion mass spectroscopy: a new
technique for the characterization of biomedical polymer surfaces. In B. D. Ratner
(ed.) Surface Characterization of Biomaterials. Amsterdam: Elsevier Science
Publishers.

42

Surface Characterization in Biomaterials and Tissue Engineering

Chinn, J. A. and Slack, S. M. 2000. Biomaterials: protein-surface interactions. In J. D.

Bronzino (ed.) The Biomedical Engineering Handbook. 2nd ed. Boca Raton, FL:
CRC Press.
Choi, E. T. and Callow, A. D. 1994. The effect of biomaterials on the host. In R. S. Greco
(ed.) Implantation Biology: The Host Response and Biomedical Devices. Boca Raton,
FL: CRC Press.
Descouts, P. A. R. 1995. Nanoscale probing of biocompatibility of materials. In A. A.
Gewirth and H. Siegenthaler (eds.) Nanoscale Probes of the Solid/Liquid Interface.
Dordrecht, The Netherlands: Kluwer Academic Publishers.
Dociu, N. and Bilge, F. H. 1986. Evaluation by scanning electron microscopy. In A. F.
von Recum (ed.) Handbook of Biomaterial Evaluation: Scientific, Technical and
Clinical Testing of Implant Materials. New York: Macmillan.
Dykstra, M. J. 1992. Biological Electron Microscopy: Theory, Techniques and
Troubleshooting. New York: Plenum.
Feng, L. and Andrade, J. D. 1994. Protein adsorption on low-temperature isotropic
carbon: I. Protein conformational change probed by differential scanning calorimetry.
J. Biomed. Materials Res., 28: 735-43.
Ferraro J. R. and Basile, L. J. (eds.) 1985. Fourier Transform Infrared Spectroscopy:
Applications to Chemical Systems, Vol. 4. Orlando, FL: Academic Press. p.61.
Fink, U. and Larson, H. P. 1979. Astronomy: planetary atmospheres. In J. R. Ferraro and
L. J. Basile (eds.) Fourier Transform Infrared Spectroscopy: Applications to
Chemical Systems, Vol. 2. New York: Academic Press.
Fischbach, F. T. 1992. A Manual of Laboratory & Diagnostic Tests. 4th ed. Philadelphia:
J. B. Lippincott.
Goodman, S. L., Li, C., Pawley, J. B., Cooper, S. L. and Albrecht, R. M. 1988. Surface
and bulk morphology of polyurethanes by electron microscopies. In B. D. Ratner
(ed.) Surface Characterization of Biomaterials. Amsterdam: Elsevier Science
Publishers.
Greisler, H. P., Gosselin, C., Ren, D., Kang, S. S. and Kim, D. U. 1996. Biointeractive
polymers and tissue engineered blood vessels. Biomaterials, 17 (3): 329-36.
Grim, W. M. and Fateley, W. G. 1984. Introduction to dispersive and interferometric
infrared spectroscopy. In T. Theophanides (ed.) Fourier Transform Infrared
Spectroscopy. Dordrecht, Holland: D. Reidel Publishing.
Hatano, M. 1986. Induced circular dichroism in biopolymer-dye systems. In S. Okamura
(ed) Advances in Polymer Science, Vol. 77. Berlin: Springer Verlag.
Hlady, V., Van Wagenen, R. A. and Andrade, J. D. 1985. Total internal reflection
intrinsic fluorescence (TIRIF) spectroscopy applied to protein adsorption. In J. D.
Andrade (ed.) Surface and Interfacial Aspects of Biomedical Polymers, Vol. 2
Protein adsorption. New York: Plenum.
Hutton B. 1999. Introduction to XPS. [Online]
http://calcium.chem.ucl.ac.uk/webstuff/people/williams/xps/xpsintro.html
Jones, D. A., Smith, C. W. and Mcintire, L. V. 1996. Leucocyte adhesion under flow
conditions: principles important in tissue engineering. Biomaterials, 17 (3): 337-47.
Knutson, K. and Lyman, D. J. 1985. Surface infrared spectroscopy. In J. D. Andrade (ed.)
Surface and Interfacial Aspects of Biomedical Polymers, Vol. 1 Surface Chemistry
and Physics. New York: Plenum.

4.6 References

43

Lee, H. B., Khang, G., Lee J. H. 2000. Polymeric biomaterials. In J. D. Bronzino (ed.)
The Biomedical Engineering Handbook. 2nd ed. Boca Raton, FL: CRC Press.
Marchant, R. E. and Wang, I. 1994. Physical and chemical aspects of biomaterials used in
humans. In R. S. Greco (ed.) Implantation Biology: The Host Response and
Biomedical Devices. Boca Raton, FL: CRC Press.
Marti, O. 1993. SXM: an introduction. In O. Marti and M. Amrein (eds.) STM and SFM
in Biology. San Diego, CA: Academic Press.
Morrissey, B. W. 1977. Adsorption of proteins at solid surfaces. Ann. N. Y. Acad. Sci.,
288: 50-64.
Norde, W. 1985. Probing protein adsorption via microcalorimetry. In J. D. Andrade (ed.)
Surface and Interfacial Aspects of Biomedical Polymers, Vol. 2 Protein adsorption.
New York: Plenum.
Palsson, B. O. and Hubbell, J. A. 2000. Tissue engineering. In J. D. Bronzino (ed.) The
Biomedical Engineering Handbook. 2nd ed. Boca Raton, FL: CRC Press.
Packer, L. 1967. Experiments in Cell Physiology. New York: Academic Press.
Pagana, K. D. and Pagana, T. J. 1997. Mosbys Diagnostic and Laboratory Test
Reference. 3rd ed. St. Louis: Mosby.
Park, J. B. 2000. Biomaterials. In J. D. Bronzino (ed.) The Biomedical Engineering
Handbook. 2nd ed. Boca Raton, FL: CRC Press.
Park, J. B. and Lakes, R. S. 1992. Biomaterials: An Introduction. 2nd ed. New York:
Plenum.
Paynter, R. W. 1988. Introduction to x-ray photoelectron spectroscopy. In B. D. Ratner
(ed.) Surface Characterization of Biomaterials. Amsterdam: Elsevier Science
Publishers B.V.
Ratner, B. D. 1988. The surface characterization of biomedical materials: how finely can
we resolve surface structures? In B. D. Ratner (ed.) Surface Characterization of
Biomaterials. Amsterdam: Elsevier Science Publishers B.V.
Ratner, B. D., Hoffman, A. S., Schoen, F. L. and Lemons, J. E. (eds.) 1996. Biomaterials
Science. San Diego: Academic Press, 1996.
Sheffield, W. D. and Matlaga, B. F. 1986. Evaluation by transmission electron
microscope. In A. F. von Recum (ed.) Handbook of Biomaterial Evaluation:
Scientific, Technical and Clinical Testing of Implant Materials. New York:
Macmillan.
Silver, S. H. 1994. Biomaterials, Medical Devices and Tissue Engineering: An Integrated
Approach. London: Chapman & Hall.
Sittinger, M., Bujia, J., Rotter, N., Reitzel, D., Minuth, W. W. and Burmester, G. R. 1996.
Tissue engineering and autologous transplant formation: practical approaches with
resorbable biomaterials and new cell culture techniques. Biomaterials, 17 (3): 237
42.
Skalak, R. and Fox, C. F. (eds.) 1988. Tissue Engineering. New York: Alan R. Liss.
Skalak, R., Fox, C. F. and Fung, B. 1988. Preface. In Skalak, R. and Fox, C. F. (eds.).
Tissue Engineering. New York: Alan R. Liss.
Spire Biomedical. Biotreatments. [Online] http://www.spirecorp.com/bio_treatments.htm
Spire Biomedical. Biotechnologies. [Online]
http://www.spirecorp.com/bio_technologies.htm

44

Surface Characterization in Biomaterials and Tissue Engineering

University of Alabama in Huntsville. Materials Science Program. Electron

Spectroscopies for Surface Characterization [Online]
http://matsci.uah.edu/CourseWare/mts723/lectures/
VG Scientific. 1999. [Online] http://www.vgscientific.com/
Wait, R. 1993. Introduction to mass spectrometry. In C. Jones, B. Mulloy and A. H.
Thomas (eds.) Spectroscopic Methods and Analyses: NMR, Mass Spectrometry and
Metalloprotein Techniques. Totowa, NJ: Humana Press.
Wintermantel, E., Mayer, J., Blum, J., Eckert, K. L., Lscher, P. and Mathey, M. 1996.
Tissue engineering scaffolds using superstructures. Biomaterials, 17 (2): 8391.

4.7 Problems
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
4.9
4.10
4.11
4.12
4.13
4.14
4.15
4.16
4.17
4.18
4.19
4.20
4.21
4.22
4.23
4.24
4.25
4.26

Search the literature to find a biomaterial listed in Table 4.1 and describe its
application.
Search the literature to find a polymer listed in Table 4.2 and describe its
application.
Describe how polymers are made more rigid.
Describe the three main structural components of cellular composites.
Search the literature to find an application listed in Table 4.3 and describe it.
State the highest practical resolution of TEM and calculate the accelerating
voltage required to achieve it. Compare with typical accelerating voltage.
Describe the principle and sample preparation for a TEM.
Describe the principle and sample preparation for a SEM.
Describe the principle and sample preparation for 2 modes of STM.
Describe the principle and sample preparation for 2 modes of SFM.
Describe the principle and sample preparation for XPS.
Describe the principle and sample preparation for AES.
Describe the principle and sample preparation for SIMS.
Describe the principle and sample preparation for ISS.
Describe the principle and sample preparation for dispersive infrared
spectroscopy.
Describe the principle and sample preparation for interferometric infrared
spectroscopy.
Describe the principle and sample preparation for FT-IR.
Describe the principle and sample preparation for ATR.
Describe the principle and sample preparation for FTIR-ATR.
Describe the principle of the contact angle method.
Describe the principle of DSC.
Describe the principle of CD.
Describe the principle of TIRF.
Describe the principle of ellipsometry.
Describe the principle of autoradiography.
Describe the principle of radiolabeling.

4.7 Problems
4.27
4.28

45

Given the data in section 4.5 and when G() = 0.37, = 100 ns, calculate particle
diameter d.
We wish to view the topography of a biomaterial surface and the molecules that
adsorb to it in a liquid. Select an instrument for imaging. Name other
instruments you rejected and why you rejected them. Describe the principle and
operation of your selected imaging instrument.

6
Cellular Measurements in Biomaterials and Tissue
Engineering
Jeffrey S. Schowalter

This chapter discusses the different types of instrumentation used in the investigation of
biological processes on the cellular level. Major advances have been made over the past
several years allowing for the observation of cells and their structure. Many of the mysteries of how various inter- and intracellular processes work have been solved in recent
years due to these advances in bioinstrumentation. One area that has benefited the most is
tissue engineering and the development of biomaterials. Tissue engineering draws from a
number of different disciplines including cell biology. Researchers are using live cells,
chemicals, and other synthetic materials to develop biomaterials for implantation in the
human body (Lewis, 1995). But a key reason researchers have been able to make this
progress is due to the significant advances in the area of cellular bioinstrumentation.

6.1 Cell Measurement Overview

Scientists have been observing cells since the invention of the microscope by Antoni von
Leewenhoek in the early seventeenth century. Microscopy, or using a microscope to
study an object, has made significant advances since then. With the advent of newer
manufacturing processes and new technology, the microscope is currently a major tool
for todays biological scientist. First lets briefly look at a basic cell, different types of
cellular measurements, and some of the parameters of interest to the biomedical researcher.
6.1.1 What is a cell?
A cell is the most basic structural unit that makes up plants, animals, and single cell organisms. It is the basic unit of life and all organisms are composed of one or more cells.

6-1

6-2

Chapter 6 Cellular Measurements

There are two types of cells: those with a nucleus (eukaryotic cells) and those without
(prokaryotic cells). Eukaryotic cells are composed of three main features: a nucleus, surrounded by cytoplasm, which in turn is surrounded by a sack-like plasma membrane.
This membrane is made from a phospholipid bilayer and is semipermeable, which means
it allows only certain things to pass in or out of the cell. Cells are 90% fluid (cytoplasm),
which consists of amino acids, proteins, glucose, and numerous other molecules. On an
elemental level, the cell contains 59% hydrogen, 24% oxygen, 11% carbon, 4% nitrogen,
and 2% others including phosphorus and sulfur. On a molecular level, the cell contains
50% protein, 15% nucleic acid, 15% carbohydrates, 10% lipids, and 10% other. Table 6.1
shows the contents of the typical cell.
Table 6.1 Typical cell content.
Structure
Cytoplasm
Organelles
Cytoskeleton
Endoplasmic Reticulum
(ER) (two types)

Description
The inside of the cell not including the organelles.
Membranous sacs within the cytoplasm.
Structural support made of microtubules, actin and intermediate filaments.
Site of protein and lipid synthesis and a transport network for molecules.

Rough ER
Smooth ER
Golgi Apparatus

Has ribosomes, and tends to be flat.

Does not have ribosomes and is tubular.
Modifies molecules and packages them into small membrane bound
sacs called vesicles.

Lysosomes
Microtubules
Mitochondria
Nucleus
Peroxisomes
Ribosomes:

Main point of digestion.

Made from tubulin and make up centrioles, cilia, cytoskeleton, etc.
Site of aerobic respiration and the major energy production center.
Location of DNA; RNA transcription.
Use oxygen to carry out catabolic reactions.
Located on the Endoplasmic Reticulum in the cytoplasm. RNA goes
here for translation into proteins.

To provide a rough idea of the scale that cell biologists and other related researchers are working on, Table 6.2 shows the typical sizes for some of the cell features.

Table 6.2 Typical sizes of cellular features

0.1 nm
0.8 nm
2 nm
4 nm
6 nm
7 to 10 nm
17 to 20 nm
25 nm
50 to 70 nm
100 nm
200 nm
200 to 500 nm
200 to 500 nm
1 m
2 m
3 m
3 to 10 m
9 m
90 m
100 m

Diameter of hydrogen atom

Amino acid
Thickness of DNA membrane
Protein
Microfilament
Cell membranes
Ribosome
Microtubule
Nuclear pore
AIDS virus
Centriole
Lysosomes
Peroxisomes
Diameter of human nerve cell
Bacteria
Mitochondrion
Nucleus
Human red blood cell
Amoeba
Human egg

Many processes that occur within cells are still not well understood. Individual
cells are complex systems by themselves but when added to their multicellular environment, the system becomes quite intricate. The human body contains over 200 types of
cells of various shapes, sizes and functions, yet each contains the same genetic material.
6.1.2 Fixed versus live cells
The study of cells by microscope can be divided into two broad categories. The traditional method of fixed cell evaluation is through the viewing of cells that have been permanently affixed to a microscope slide. The other way to examine cells is while they are
living (e.g. as part of a tissue or in a controlled environment like a petri dish).
Fixed cells
In some cases, with relatively loose or transparent tissue, the object under examination
can be placed on a slide and then placed under the microscope. If the item of interest is
part of a larger object, such as an organ, or a bone, then the tissue must be sliced. If the
object is generally opaque, then the slice either must be very thin or some other method
must be used as discussed in Chapter 4. Structures inside most cells are almost colorless
and transparent to light. Therefore, dyes are used which react with some of the cells
structures but not others. Staining the cells may alter the structure of the cell or kill it.
Because of this, the cells are normally preserved or fixed for observation. This is normally done with some fixing agent such as formaldehyde. In fixing the cells, the general
structure of the cell and organelles may be preserved but enzymes and antigens within the
cell may be destroyed. Usually the fixing process takes at least 24 h.

6-4

Chapter 6 Cellular Measurements

Another method involves the freezing or cryofixing of cells for observation.

Typically a section of tissue is quick frozen to 15 to 20 C using liquid nitrogen and is
sectioned in a device called a cryostat. A cryostat is a microtome mounted in a freezing
chamber. Thin sections can be prepared in minutes and the sections usually retain a frozen snapshot of their enzymes and antigens. However, this method has the disadvantage
that ice crystals formed during the freezing process may distort the image of the cell and
the specimens have a tendency to freeze-dry. Also, the sections must be used immediately. This is good when trying to make a diagnosis while a patient is on the operating
table but not so good for long-term research studies.
To get some semblance of time responses using fixed cells, a procedure is used
where a stimulus is provided to the cells and then they are quick frozen moments later.
The characteristics of individual cells at different stages of response to the stimulus are
examined and then these individual responses are pieced together into one response that
describes the entire process.
Sometimes the researcher may want to isolate a certain component of the cell for
study. The process of breaking open cells is called homogenization and the separation of
organelles is called fractionation. Isolating organelles requires techniques such as gravity
sedimentation and centrifugation.
Homogenization is normally accomplished by using force to break the cell
membrane. The most common procedures use glass mortar and pestle arrangements with
controlled bore sizes. When the pestle is inserted and turned rapidly in the tube containing small pieces of tissue, intracellular connections are broken and membranes of whole
cells are ruptured. The resulting mixture is a suspension of organelles and other material
from the broken cells. Ultrasonic waves are also used where the cell membrane is broken,
but the organelles remain intact.
After homogenization, the various components must be separated. For some
materials, this can be accomplished by simply using gravity. Gravity separates the organelles naturally due to differences in their density. The most widely used instrument for
separation, however, is the centrifuge. In centrifugation, a mixture of cell components is
subjected to increasing centrifugal forces. At each level, the material which sediments
can be saved and the supernatant can be centrifuged at higher forces. Organelles of low
density and small size sediment at high forces, while organelles of high density and large
size sediment at low forces. A centrifuge that operates at speeds greater than 20,000 rpm
is called an ultracentrifuge. Speeds can reach 110,000 rpm yielding 199,000 g.
Live cells
Live cells have the advantage that time is not typically a factor when viewing. The cells
are viewed in their natural environment and time studies are easily accomplished using
video or other recording devices.
One of the three general categories of live cell research is the division of cell
population studies. This is done while the cells are free to interact with one another. With
this configuration, not only the activity within the cell can be studied, but also the interaction between cells and other substances or objects external to the cellular structure such
as man-made materials.

A second category of cell research involves examining the characteristics of an

individual cell. This may be the physical features of the cell, its topology, or the relative
chemical/molecular structure of the cell. Another area of study involves the response of
the cell to experimental treatments.
Single cell imaging involves one of two different techniques. The first is the 3-D
reconstruction of a scanned object. Using digital (computer) techniques, characterization
of single cells can be done by the cell researcher. The other technique involves time sequencing of the movement, or morphology, of a cell. This involves the use of video cameras and video recording systems. For example, cell adhesion may be studied by video
and fluorescence imaging of the spreading and detachment of adherent cells subjected to
a range of fluid shear stresses in a radial flow chamber. Cell motility is assessed by timelapse video microscopy of individual cells crawling on sheets of polydimethylsiloxane.

6-6

Chapter 6 Cellular Measurements

6.2 Light Microscopy

Since there is a theoretical limit (Rayleigh limit) on how much an object can be magnified and still be seen clearly, scientists have resorted to a variety of new technologies and
have developed a number of techniques to enhance their understanding of the cellular
world.
Light microscopes rely on the bending, or refraction, of light. When a curved
lens is used, light rays pass straight through the center of the lens while rays passing
through the outer edges of the lens are bent. The farther away from the center light passes,
the more the bending, as shown in Figure 6.1(a). All light rays pass through one point, the
focal point, F, placed at the focal length LF.
Figure 6.1(b) shows that when an object is placed outside the focal length LF, it
is magnified by the magnification mm = LI/LO where LI is the image length and LO is the
object length and lengths are related by the lens makers' equation

1
1
1
+
=
LO LI LF

(6.1)

Object
Image
F

LO
LF
(a)

LI

(b)

Figure 6.1 (a) Light rays through a lens and the corresponding focal point F. (b) The light rays are
bent by the lens and refocus as an image.

Figure 6.2 shows that a conventional (wide field and bright field) compound
microscope is composed of two lenses. The primary lens (objective) has a very short focal length and is placed close to the sample. It forms a much larger primary (intermediate)
image just below the eyepiece. The secondary lens (eyepiece) alters the primary image by
forming parallel rays that are focused on the retina by the lens of the eye. This is the
enlarged virtual image seen when looking into a microscope. Light is intensified by the
condenser lens so it more readily transmits through the sample.

Eye

Eyepiece

Objective
Slide

Sample

Condenser
Field
iris
Light
source
Figure 6.2 Compound microscope. The compound microscope contains a light source for sample
illumination, a field iris to control the light field, a condenser to focus the illuminating light, an
objective lens, and an eyepiece.

If an electron beam replaces the beam of light, the microscope becomes a transmission electron microscope (Chapter 4). If light is reflected from the object instead of
passing through, the light microscope becomes a dissecting microscope. If electrons are
reflected from the object in a scanned pattern, the instrument becomes a scanning electron microscope (Chapter 4).
6.2.1 Resolution versus magnification

The microscopes function is to allow viewing of objects that cannot be seen with the
human eye. Because we are trying to make an object look larger, the term resolution is

6-8

Chapter 6 Cellular Measurements

often confused with magnification. As previously discussed in Chapter 4, magnification

refers to the size of the image as compared to the original object and is usually expressed
as mm, where mm is the amount of magnification. However, resolution determines
whether small objects that are close together can be distinguished as separate objects. As
a general rule, the greater the magnification, the greater the resolution. However, there
are several practical limitations of lens design that can result in increased magnification
without increased resolution. If an image of a cell is magnified by a factor of ten, the image will get larger, but not necessarily any clearer. Without resolution, no matter how
much the image is magnified, the amount of observable detail is fixed. Regardless of how
much you increase the size of the image, no more detail can be seen. At this point, you
have reached the limit of resolution. This is also known as the resolving power of the lens
and is only a function of the lens.
The human eye can only distinguish between two objects when they are about
0.1 mm apart at a distance of 100 mm. If two objects are 0.01 mm apart, they cannot be
distinguished unless we magnify their image by 10. In this case, the limit of resolution
has changed from 0.1 mm to 0.01 mm, or inversely, our resolving power has increased by
a factor of 10.
Unfortunately, a lens can magnify an image without increasing the resolution.
Lenses are not perfect and cause objects to become blurry at the edges. Thus, even
though they can be made to appear 0.1 mm apart, the edges are so blurry that we lose the
ability to see them as two distinct objects. While microscope lenses are typically expressed in terms of magnification, the important value is the resolution.
The resolution of a lens is a function of how it is configured and the wavelength
of light that is passed through the lens.

resolution = 0.61
n sin

(6.2)

where is the wavelength of the illuminating light, n is the refractive index of the medium between the objective and the specimen, and the cone angle is 1/2 the angle of the
acceptance of light onto the lens (Figure 6.3). The value n sin is known as the numerical
aperture (n.a.).
Objective

Specimen
Figure 6.3 Cone angle .

There are three ways to improve resolution. One is to use shorter wavelengths of
light. This is typically done using filters. Another is to increase the cone angle. The cone
angle can be increased by placing the objective as close as possible to the specimen. Practically, the largest value of sin is 0.95 (GEK, 1997). Finally, the refractive index, n, can

be increased. Immersion oil is sometimes used in place of the air between the objective
and the slide to increase n.
With bright field microscopy, work is usually done in the visible light range so
the image can be seen directly with the human eye. Since the shortest wavelength of visible light is blue, almost all microscopes have incorporated a blue filter, which is often
referred to as a daylight filter. However, the human eye has a maximum sensitivity to
green and since specimens contain various colors, the resolution of an image typically
varies between 0.2 to 0.35 m depending on the color of the light source. Resolution can
be enhanced by reducing the wavelength to the ultraviolet range. However, ultraviolet
(UV) light cannot be seen directly by the human eye. Thus, this form of microscopy requires a detector or photographic film that is sensitive to UV. UV light improves the resolution by a factor of two over visible light.
The best resolution is not attainable, however, unless the lenses are corrected for
problems common to lens design. Modern microscopes normally use a set of lenses assembled to maximize the refractive index while minimizing chromatic and spherical distortions. Chromatic and spherical distortions (aberrations) occur in all lenses to some
extent due to the fact that a 3-D object is being viewed as a 2-D image. However, aberrations are typically determined by how well the lenses were designed and manufactured.
In general, the smaller the optical distortion, the greater the cost.
6.2.2 Light microscope modes
Bright field

The light microscope can use several methods to view objects. Bright field microscopy is
used to view stained or naturally pigmented organelles in the cell. It involves the use of
color detection. If an object contains colored features, the eye detects those colors. The
stains or pigments absorb and alter the light that passes through the sample. Therefore,
the light that passes through the sample is seen as a color that is distinctly different from
the background or direct light that does not pass through the sample.
Most of the features in the cell are relatively colorless and difficult to see under the microscope. The easiest way to enhance the contrast of the relatively transparent cells is to
stain them. Different colored dyes are used that stain different parts of the cell. If the cells
are fixed, this is not a problem except for possible damage to the cellular integrity from
the dye.
Further image enhancement can be accomplished by using a specific monochromatic light source. One way to achieve this is by using a filter. The most commonly
used filter is one that allows only blue light to pass through. Blue light, about 450 nm, is
usually the shortest wavelength used in light microscopy. It determines what cell features
you can and cannot see under the microscope. In order to see a specific structure of the
cell, it must be large enough to be able to perturb the wave motion of the light rays that
strike it. Therefore, mitochondria, about 3 m, are one of the smallest organelles to be
seen through microscopy. Organelles that are smaller than this cannot be resolved be-

6-10

Chapter 6 Cellular Measurements

cause they are small compared to the wavelengths of visible light and are not able to perturb the wave motion of light traveling to our eyes.
Phase contrast

Another method of viewing objects is called phase contrast microscopy. This type of microscopy is used to view living cellular structures because they are not pigmented, thus
unable to be seen under bright field microscopy, and the structures within the cell tend to
differ in refractive indices.
Phase contrast takes advantage of the fact that when rays of light pass through
areas of differing refractive index or different lengths of optical path (thicker sections),
the light rays get out of phase with one another. This causes interference, which in turn
causes different areas to appear as differences in brightness when seen by the eye. Phase
contrast microscopy makes use of these minute shifts in the phase as the light passes
through an object. If the normal phase shift is increased by 1/4 or 1/2 wavelength, the
differences will be magnified in intensity. Phase contrast microscopes that provide 1/4
wavelength shift are called medium phase while those that provide 1/2 wavelength shift
are called dark phase.
A standard compound microscope is modified for phase contrast microscopy, as
shown in Figure 6.4, by adding annular transparent rings, one below the eyepiece and
another below the condenser. Some light transmitted from the lower annular ring passes
through the specimen and is imaged on an upper semitransparent ring. Other light transmitted from the lower annular ring is refracted slightly by different parts of the specimen
and passes outside the upper semitransparent ring and is imaged on a surrounding transparent diaphragm, which is one-quarter wavelength thicker. Since the rays of light travel
slightly different distances, when they meet up again, they will be slightly out of phase
with one another. If the phase differences add to each other, constructive interference, the
image will be brighter (Figure 6.5). If they cancel out, destructive interference, the image
will be darker (Figure 6.6).

Upper annular
phase
changing ring
Deflected
light

Objective
Sample
Slide

Condenser

Lower
annular ring

Figure 6.4 In phase contrast microscopy, light from the lower annular ring is imaged on a semitransparent upper annular ring. This microscope uses the small changes in phase through the sample to enhance the contrast of the image.

Figure 6.5 The interaction of light waves that meet in phase results in constructive interference.
The amplitude of the wave is doubled by this interaction.

6-12

Chapter 6 Cellular Measurements

Figure 6.6 Interaction of light waves that meet one-half wavelength out of phase results in destructive interference. The waves cancel each other when they intersect.

If the upper annular ring is opaque instead of semitransparent, it will block all
light except that which is refracted, thus yielding a dark field phase contrast microscope.
Differential interference contrast (DIC)

DIC imaging is a modification of phase microscopy and also produces contrast from differences in refractive index in the sample. Contrast is obtained by converting these phase
changes into amplitude changes. DIC does this by splitting light with a prism into two
parallel rays. These rays are routed through two different optical paths and then recombined. As with the phase contrast technique, the rays interfere with one another constructively or destructively depending on the phase difference. The advantage to this method is
that the phase difference can be shifted in phase by a variable amount. The most commonly used configuration of differential interference microscopy is called the Normarski
Interference Microscope, named for the optical theoretician George Normarski. In microscopy, the term Normarski is synonymous with the differential interference contrast
technique. This technique is used for measuring thickness of cells and observing other
items with little or no contrast. DIC imaging provides good rejection of out-of-focus interference. Essentially it provides an image of the rate of change of refractive index in the
sample. The technique essentially acts as a filter that emphasizes edges and lines. Out-offocus refractive index changes will be blurred and have a small rate of change in the focal
plane so they will not contribute much to the contrast of the image.
Dark field

Figure 6.7 shows another method called dark field microscopy. With dark field illumination, the background field is dark because the condenser light comes from outside the
collecting angle of the objective. This is due to an opaque stop below the condenser.
Small objects, even those below the limits of resolution, can be detected easily since objects scatter light and appear to emit light on a dark field.

(a)

(b)

Figure 6.7 Example of darkfield illumination. (a) shows the typical brightfield view while (b)
shows the darkfield view of the same object. Imagine the object in the darkfield view glowing like
the moon at night.

Rheinberg illumination, also called optical staining, is especially useful for looking at unstained samples. This method is similar to conventional darkfield illumination
except that the central opaque light stop at the base of the substage condenser is substituted for a colored central stop. A sample viewed by this method looks white on a background that is the color of the stop. If a colored transparent annular filter is added, the
sample is colored.
Inverted microscope

If a light microscope is functionally turned upside down to the object and viewed from
under the sample, the microscope is called an inverted microscope. The light source and
condenser are above the stage pointing downward. The objective is below the stage pointing upward. As a result, one is looking up through the bottom of the specimen and rather
than looking at the specimen from the top, typically through a cover glass, as on a conventional microscope. This is particularly useful with live cells since they can be seen
through the bottom of a container, without opening the container, and without the air interface normally present between the objective and the surface of the culture. By adding
phase contrast capabilities, cells can be observed without stains.
Common biological stereo-binocular (dissecting) microscopes are also used with
reflected light to examine tissue that is too thick to allow light to pass through it.
Near field scanning optical microscopy (NSOM)

NSOM offers dramatic improvement in resolution from about 1 m to about 10 nm

(Paesler and Moyer, 1996). The lens, which has a far field diffraction limit, is replaced by
a tapered optical fiber probe with an apex of curvature of about 10 nm. The sample is
raster scanned in close proximity to the fixed probe, which detects light only from the
surface in the immediate vicinity. The sample is placed on a vertical piezoelectric tube.
Electrodes on the left and right sides are driven with 200 V to achieve 15 m side movement. Electrodes on the front and back permit forward and backward movement. An interior electrode permits 1.7 m of vertical movement.

6-14

Chapter 6 Cellular Measurements

6.2.3 Electronic microscopy

A photographic, video, or solid-state camera can be attached to a microscope to make a

permanent record of an observation. There are three types of microscope viewing ports:
monocular (one eyepiece), binocular (two eyepieces), and trinocular (two eyepieces and a
photography port). A camera can be adapted to each of these viewing heads. Camera
adapters usually are attached to one of the viewing ports and adjusted to be parallel with
the eyepieces. The microscope itself usually acts as the camera lens.
For many years, researchers used only still cameras or video camera systems to
acquire light microscope images (Shotton, 1993). The use of a still camera is called photomicroscopy and the still picture is called the micrograph. The use of a video camera is
called video microscopy. The most recent instrumentation involves the use of specially
configured microscopes, new types of high-resolution solid-state cameras, image digitization, and digital image processing. Some of these techniques and methods were the result
of technological advances while others, most notably video enhanced contrast microscopy, were discovered by accident. However, as a result of these developments, the field
is now usually called electronic microscopy or optodigital microscopy (Shotten, 1990).
Solid state cameras

Solid-state cameras are ideally suited for many cellular microscope applications. The
typical solid-state camera uses a technology called a charge-coupled device or CCD.
These devices consist of a large square array of tiny detectors (called pixels) each of
which produces an electric charge in response to incident photons (see Figure 6.8). Each
pixel is about 20 m2. The charge for each pixel is collected and sent to an output amplifier so that the level of light incident to each pixel is recorded. Charge accumulates over
time so the total amount of charge is proportional to the product of the light intensity and
the exposure time. The intensity of each pixel is displayed as a video matrix on the computer, yielding a two-dimensional image of incident light. Tiny red, green and blue filters
placed in front of sequential pixels yield a color camera.
The CCD camera has very high signal-to-noise ratio and is available in high
resolution arrays of 2032 2044 pixels or greater at 1.6 frames/s. These array cameras
have a response in which the output is linearly related to the input light intensity over a
very wide range of light levels. In addition, they are not susceptible to accidental light
damage as are the high sensitivity video cameras. A lower resolution array of 1300
1030 pixels at 12 frames/s is still below the 60 Hz standard video scan rate. This limits
these devices to studying either static or slow dynamic processes. Also, many times these
devices are operated at low temperatures to maximize the signal-to-noise ratio and are
therefore termed cooled.

Single pixel

CCD array

Figure 6.8 CCD Array. The CCD array is made up of many light-sensing elements called pixels.
Each pixel can independently sense light level and provide this information to a computer.

Wide field versus point scan imaging

It is important to distinguish between the two types of image acquisition (see Figure 6.9)
that can be done with these microscopes. Image plane scanning, or wide field imaging,
involves the acquisition of a two dimensional image by the recording device. The device
is focused on a plane of the object and a two-dimensional picture is directly recorded.
This is the type of image that you would see if you looked into a traditional light microscope. In contrast, a point-scanned device is focused on a single point of the object. The
instrumentation must have some way of scanning or focusing on different points. This is
typically done in the two horizontal (xy) planes so the resulting picture would look the
same as the wide field image (Shotten, 1990). An example of this is the scanning electron
microscope discussed in Chapter 4.
Objective

Objective

Sample

Sample

(a) Wide field

(b) Point scan

Figure 6.9 Comparison of the wide field versus the point scan techniques. (a) wide field collects an
entire image while (b) point scan image must be assembled point by point.

6.2.4 Cellular markers

6-16

Chapter 6 Cellular Measurements

There are two general methods for highlighting or marking different structures within a
cell or to mark different types of cells. These markers are used for a variety of purposes
from locating a certain structure within a cell, identifying which cells contain a particular
substance, how much of the substance they contain, where it is located within the cell,
how that substance may move over the cells cycle, if the substance moves between cells,
and even how much of the substance is present within a cell.
These markers can be put into the cells in several different ways. Some are naturally taken up by the cells when introduced into their environment. Others are injected
into cells using special techniques or devices such as the tiny syringe-like microinjector.
Once put into cells, some are analyzed immediately, others are given some time to bond
to specific cellular structures, while others are activated at a later time when the researcher wants to observe time specific cellular dynamics.
Fluorescent probes

The use of fluorescent markers or probes is revolutionizing the imaging of live cells and
their associated contents, structures, and dynamics. These systems all involve the excitation of a fluorescent material, the subsequent fluorescence emission by the material, and
the detection of this fluorescence by a detector (Haugland, 1997).
The probes, also called fluorochromes and fluorophores, are fluorescent materials that are introduced into the cellular environment. There are a number of different
ways that these probes are used with the cells:

the probe binds directly to the cellular structure of interest,

the probe binds to another molecule that has an affinity or reactivity towards
some cellular structure,
the probe binds to an antibody that in turn binds to some cellular structure,
the probe does not bind but its fluorescence depends on local concentration of a
solute,
nonfluorescent molecules are converted to fluorescent molecules through an enzymatic reaction.

The use of these probes is especially useful for microscopy for several reasons.
First the excitation wavelength is different from the emission wavelength. This allows for
easy detection because the excitation wavelength can easily be filtered out. Also, different types of probes emit fluorescence at different wavelengths. This too is especially useful in that several probes can be used to mark different structures within a cell and these
can be detected at the same time. In addition, the entire excitation/fluorescence cycle can
be repeated.
Two considerations are important regarding probe selection. First, probes must
be chosen based on sensitivity, or the amount of substance that is labeled within the cell.
Also the probes specificity or the ability to only label the substance of interest is important. Specificity is provided by fluorochromes. These are either synthetic chemicals or
natural products that can have innate specificity, binding to certain classes of molecules,
or they can be attached to molecules that have specificity (for example, antibodies). Some

types of fluorochromes are generated by the activity of enzymes, or are sensitive to the
presence of specific ions. A final class of fluorochrome is generated by gene expression.
The advantages of using probes include the high signal levels against a low
background (noise) level, they are not prone to motion artifact, and they can be used to
monitor very small changes.
Care must be taken regarding the intensity level of the light used to excite the
probe. Photobleaching is the breakdown of the fluorescent probe within the sample as it
is exposed to light. Typically, the more intense the light, the more photobleaching that
occurs. Photobleaching causes several problems. It reduces the fluorescent signal emitted,
it generates oxygen free radicals, which can damage cells, and it may generate forms of
dye that distort the relationship between the substance being marked and other areas of
the cell.
In a flow cytometer, a laser excites fluorescence of individual cells in an interrogation volume. The fluorescence is then separated on the basis of color, through the use
of optical filters, and the results displayed as a histogram of cell count vs. cell type. Cells
may also be selectively isolated or sorted to high purity as they pass through this system.
Fluorescence video microscopy can capture multiple images at differing wavelengths of light simultaneously at video rates. The fluorescence event follows exponential
kinetics with a lifetime of approximately 350 ms.
Radioactive tracers

Radioactive tracers are used in a number of areas of cell research as an effective and safe
means of monitoring cellular interactions. This was the primary technique used for years
but is being used less and less with the development of more and better fluorescent
probes. These tracers have the advantages of high precision and specificity but suffer the
disadvantages of extensive radiation handling safety procedures, the inability to distinguish more than one probe at a time, and a relatively long time (hours to days) to obtain
results.
Radioactivity is caused by the spontaneous release of either a particle and/or
electromagnetic energy from the nucleus of an atom. Tritium (3H) is an important tracer
used for cellular research. Other radioactive elements used include 14C (Carbon-14), 32P
(Phosphorus-32), 125I (Iodine-125). Tritium is most prominent because replacing hydrogen with tritium in a molecule does not cause chemical or physiological changes in molecular activity (Heidecamp, 1996).
Radioactive elements emit radiation spontaneously, but over time a percentage
of all radioactive elements in a solution will decay. It is nearly impossible to predict
which individual element will radioactively decay, but predictions can be made about
large numbers of the elements. The radioactivity of a material is the rate of how many
decays occur each second. It is measured in becquerels.
When an alpha or beta particle, or a gamma ray passes through a substance, ions
are formed. To monitor this ionization effect, a device called an ionization chamber is
used. The most common forms of ionization chambers are the Geiger counter and the
pocket dosimeter. These chambers have two electric plates separated by an inert gas with
a voltage applied across them with a battery or power supply. When a particle enters the

6-18

Chapter 6 Cellular Measurements

chamber, ions form and are drawn to one of the electric plates as shown in Figure 6.10.
The negative ions attract to the positive plate (or anode) and the positive ions attract to
the negative plate (or cathode). This causes a small voltage change in the form of a pulse
to temporarily occur across the plates. This pulse is passed by the capacitor, which blocks
the dc voltage from the power supply. The output pulse is then read by some measuring
device. The sensitivity of the system depends on the voltage applied between the electric
plates. High-energy alpha particles are significantly easier to detect than beta particles,
and they require lower voltage. In addition, alpha particles penetrate through the metal
casing of the counter tube, whereas beta particles can only pass through a quartz window
on the tube. Consequently, ionization chambers are most useful for measuring alpha
emissions. High-energy beta emissions can be measured if the tube is equipped with a
thin quartz window and if the distance between the source of emission and the tube is
small.
Resistor

+
Voltage
supply

Capacitor

+++++++++++++++
Particle

+ +
Ions
+

Pulse
count
Plates
Freq to volt
counter

Meter

Figure 6.10 Basic radiation counting system. The power supply has a voltage source and series
resistor. Ionization by a radiation particle causes a voltage pulse, which passes through the capacitor and is read by the meter.

A pocket dosimeter is a modification of the basic ionization chamber. It uses a

miniaturized tube that can be charged to hold a voltage without constantly recharging the
unit with a power source. Essentially it is just a capacitor, which is charged by a base unit
and carried as a portable unit. It is small enough to be easily carried in a pocket. When
exposed to ionizing radiation, the capacitor discharges slightly. Over a period of time, the
charge remaining on the dosimeter can be measured and radiation exposure can be calculated. Since dosimeters carry lower voltage potentials, they are used for the measurement
of X-ray and high-energy gamma radiation, but do not detect beta emissions.
Liquid scintillation counters are also used to measure radiation. They use a substance in the counting chamber, called a scintillant, to produce a flash of light. This flash
is produced for each radiation particle detected in the chamber. A circuit then converts
each light flash to an electric current pulse. This signal is then sent to a counting system,

which keeps track of total counts or counts per minute (cpm). If the efficiency of the system (or the percent of actual radioactive decays that were detected) is known, the disintegrations per minute (dpm) can be calculated using Eq. (6.3).
dpm = cpm/efficiency

(6.3)

Autoradiography is the process of using radioactivity to trace cellular activity.

Tritium is used in cell analysis because it is relatively weak and can be localized within
cell organelles. Carbon and phosphorous are also used, but are more radioactive so they
require more handling precautions and provide poorer resolution of intracellular details.
They are more commonly used to analyze details at the tissue or organ level.
The picture of a cell on photographic film that has been labeled with radioactive
tracers is called an autoradiogram. When silver ions in a photographic emulsion are exposed to radiation they change to silver grains. The number and distribution of these
grains are therefore proportional to the amount and distribution of the cellular radioactive
label.
The quantity of silver grains depends on the type of emulsion and the kind of
radiation emitted from the cell. Alpha particles produce straight, dense tracks a few micrometers in length. Gamma rays produce long random tracks of grains, which are useless for autoradiograms. Beta particles or electrons produce single grains or tracks of
grains. High-energy beta particles (32P) may travel more than a millimeter before producing a grain. Low-energy beta particles (3H and 14C) produce silver grains within a few
micrometers of the radioactive area thus providing excellent resolution.
6.2.5 Confocal laser scanning microscopy

Confocal laser scanning microscopes (CLSMs) have been a significant innovation for the
light microscope industry (Wright, 1993). These microscopes remove the out-of-focus
blur from the 2-D image directly through hardware rather than requiring a software deblurring technique (discussed in section 6.4.1). In addition, this technique yields images
from relatively thick biological tissue and successive 2-D optical sections (known as a Zseries) can be assembled in an imaging system to construct a 3-D image. A standard light
microscope can be used for CLSM with some technically simple, yet expensive, modifications.
As opposed to a conventional wide field microscope which produces a 2-D image directly, the CLSM is a point scan device, which builds a 2-D image by scanning and
assembling the image point by point with an integrated imaging system. A CLSM normally uses a laser beam as a light source because a laser provides high intensity, wellfocused, monochromatic light, which is able to easily illuminate individual points within
a specimen. Using a point light source avoids the interference of scattered light when the
entire field is illuminated. This interference is known as halation.
The scanning is typically accomplished by a series of xy plane rotating mirrors
in the light path between the dichroic mirror and the objective that focus the laser beam
on different points of the specimen in the xy plane. To obtain an image of the specimen,
the light from the specimen is routed to a detector. A pinhole allows only the light origi-

6-20

Chapter 6 Cellular Measurements

nating from the focal plane to reach the detector as shown in Figure 6.11. The detector
(usually a CCD camera or photomultiplier tube) measures the intensity of light and produces a signal proportional to the light at each point. This signal is then typically sent to
an imaging system where the current is converted to a voltage, digitized, processed to
reduce noise, and an image is reconstructed for display on a monitor.
CLSM has several advantages over conventional microscopy. CLSM provides
high-contrast images of specimens without the image artifacts normally present with
conventional contrasting techniques. It also has the ability to view relatively thick specimens and is well suited for fluorescence applications as will be discussed later. The disadvantages of this technique are relatively high cost and possible damage to the specimen
from the laser illumination. In an attempt to minimize this damage, each slice may be
scanned several times at relatively low laser power and the results integrated.
Detector
Pinhole
detector
aperture
Pinhole
illumination
aperture
or laser

Dichroic
mirror
Objective
Specimen

Focal plane

Out of focus
In focus
Figure 6.11 Confocal laser scanning microscope. The microscope removes out-of-focus (z-plane)
blur by keeping out of focus light from reaching the detector using a pinhole.

6.2.6 Two-photon excitation microscopy

Another technique that has been developed to minimize the excitation lasers potential
damaging effects is two-photon excitation microscopy (TPEM). This technique is based
on the principle that two photons of longer wavelength light simultaneously excite a
probe that would normally be excited by one photon of a shorter wavelength, as shown in
Figure 6.12. This method has the advantage that less photobleaching occurs because only
the focal point is being excited as compared to the entire depth of the specimen as in confocal and wide-field imaging. (LOCI, 2000). In addition, TPEM can obtain optical sections from deeper into a specimen. This is possible because the excitation light is not absorbed by the probe above the plane of focus and the longer wavelengths are scattered
less within the sample. Two-photon excitation has several limitations. A given probe has
lower resolution when compared to confocal imaging. This can be eliminated by the use
of a confocal aperture at the expense of a loss in signal. Also, thermal damage may occur
if the specimen contains substances that absorb the excitation wavelengths.

Detector

Two beam excitation

Dichroic
mirror

Objective

Specimen
Focal plane
Figure 6.12 Two photon excitation microscope. This microscope doesnt require a pinhole but is
able to excite single points within the sample by having two photons excite the sample only at the
exact location of interest.

6-22

Chapter 6 Cellular Measurements

6.2.7 Image processing

The advent of faster and cheaper personal computers has permitted a myriad of image
processing techniques that seem ideally suited for the microscope. Several of these techniques correct the image in software as opposed to hardware. Others take the large volumes of data generated by digital imaging and manipulate it into a form that is more readily usable by the researcher.
Computational deblurring

A computational deblurring (digital deconvolution) light microscope system is a relatively new system that offers an alternative to confocal microscopy. These types of systems use a computer interfaced with a conventional light microscope to remove the blur
(haze) from the 2-D image by removing the effects of out-of-focus light from above and
below the plane of interest. By a fairly complex mathematical algorithm, each 2-D slice
of the object is stored in the computer as a digital image (Richardson, 1997). This image
is then processed with an imaging algorithm. It mathematically makes a first estimate as
to which 2-D plane light has originated from based on the characteristics of the lens system used. The computer then removes the out-of-focus effects from neighboring slices.
Then the adjusted image is compared to the original image and the error is determined
and used to adjust the estimate. This process is iteratively repeated until a minimal error
is obtained. High-resolution fluorescence images are achieved. This method also enhances detection of low-level fluorescence. Image processing and analysis is typically
done after imaging has been completed although some systems process the data real-time.
Three dimensional cellular tomography

Three-dimensional cellular tomography is a process for making visually-traced, contourbased 3-D reconstructions and measurements from microscopes. This process is similar
to the technique used for X-ray and magnetic resonance imaging discussed in Chapter 7.
These systems process images from TEMs, SEMs, confocal microscopes, and light microscopes using stacks of vertically aligned 2-D sections. An algorithm fills in structures
and builds a single 3-D image. Several thin sections are combined into artificial thick
sections. Tracing can be done within the entire volume or in a series of overlapping thick
sections. After tracing, contours are combined and displayed as reconstructed objects
using various orientation and display options (stereo, color, rotation, surface rendering,
etc.). The contour data are also assigned 3-D coordinates to calculate volumes, surface
areas, and perimeters. In addition, low-contrast structures become more apparent and
better defined when looking through a volume rather than at individual slices. Manipulations can be done to create a stereoscopic image of the image so it can be viewed in 3-D.

6.3 Cell Orientation

Cell orientation refers to both the orientation of cells within tissue and the orientation of
cellular components within the cell. There are a number of reasons to measure the orientation of cells within tissue. Understanding how cells are oriented in relation to one another, and thereby the mechanical properties of natural tissue, is an important aspect in
the development of artificial tissue (Tranquillo, 1997). Many types of cells have been
found to have specific spatial orientations where the top and bottom of a cell can be identified. Another aspect involves the way in which cells orient themselves and migrate in
response to different types of stimuli. An understanding of this process will aid in the
growth of artificial tissue.
6.3.1 Orientation chamber

One way to measure orientation is by using an orientation chamber. One type of orientation chamber tests the ability of cells to orient themselves in a gradient of chemical
attractant. The chamber is similar to a hemocytometer (see Chapter 5). The gradient is set
up by diffusion from one well to the other and the orientation of cells toward the well
containing chemical attractant is scored on the basis of their morphology or by filming
their movement.
6.3.2 Video enhanced contrast microscopy

Video enhanced contrast microscopy (VECM), as shown in Figure 6.13, uses a highresolution video camera and a light microscope that has differential interference contrast
(DIC) optics. By using both the analog and digital contrast enhancement as well as background image subtraction, this method is able to observe objects with dimensions an order of magnitude smaller than the resolution limits of the microscope alone (Shotten,
1990). During image subtraction, for example, all identical structures in successive video
images can be erased, filtering out any interfering background images. This method involves three steps (Figure 6.14). First the light microscopes DIC optics is adjusted to just
under the saturation level of the video camera. The second step involves reducing the
background level by adjusting the black-level control. Then the gain control is adjusted to
amplify this dark-low intensity signal. The final step is required because the previous one
also captures background blemishes created by lens imperfections and unevenness of
illumination. In this step, the specimen free background that has been previously recorded
is subtracted from the live image.

6-23

6-24

Chapter 6 Cellular Measurements

Video
camera

Frame grabber

Light
microscope
Monitor and
computer

Video
recorder

Figure 6.13 Video-enhanced contrast microscope (VECM) system. A light microscope image is
recorded by a camera and the image is sent to a recorder and a computer for image processing.
Intensity

Intensity

Distance

Intensity

Distance

Intensity

Distance

Distance

Figure 6.14 (a) Original low-contrast video image. (b) The threshold is set just under the saturation
level. (c) Reducing the background level. (d) Gain is adjusted to amplify dark-low intensity signal.

6.3 Cell Orientation

25

Figure 6.15 (a) An unprocessed photo of cells of the inner epidermis taken through an interference
contrast microscope. (b) The same image with digital contrast enhancement, the single structures
become apparent. The background fault remains. (c) Subtraction of the background and resulting
with further contrast enhancement.

6.4 Cell Rolling Velocity

Cell rolling velocity involves the tracking of cells, particularly blood cells, as they start to
tumble or roll as the body responds to infection or foreign matter. This is a parameter that
becomes important in the areas of healing and the bodys response to biomaterials implanted in the body. When an infection occurs in an area of the body, white blood cells
are responsible for attacking the infection. The way the body signals these cells is critical
to understanding the healing process. Recruitment of white blood cells from the blood to
the area of infection is required. This recruitment comes from cell adhesive molecules on
the inside of the blood vessel. These Velcro-like cell adhesion molecules (CAMs) start
the rolling process and cause the white blood cells to roll along the endothelium wall of
the blood vessel. Since this is the first step in the process of the activation of the bodys
immune system it has application in the response to tissue engineering and biomaterials.
This is typically measured using fluorescent probes, intensified fluorescence microscopy,
cell tracking techniques, and the artificial introduction of CAMs.
Intensified fluorescence microscopy (IFM) involves the use of a fluorescent
specimen, excitation of the specimen with low-level light, and detection of the fluorescence with a microscope and a high sensitivity video camera. High-sensitivity video
cameras use video image tubes with sensitivities about two orders of magnitude greater
than conventional video cameras. This configuration was known as video-intensified
fluorescence microscopy (VIFM). Recently, these cameras have been replaced with solidstate slow-scan cooled CCD cameras of high resolution. These devices are ideally suited
for low-level light imaging and do not suffer from many of the problems associated with
the high-intensity video cameras. IFM images do have one major problemout-of-focus
blur. Since the entire field of view of the specimen is illuminated, there are fluorescent
emissions from the areas above and below the focal plane causing the out-of-focus blur.
This problem is compensated for in software and is detailed in the image processing section 6.2.7. The actual computation of rolling velocities is typically accomplished after
imaging has been completed by using cell-tracking software, which computes the rolling
velocities from relative probe locations from the prerecorded images.
Figure 6.16 shows a block diagram of an IFM system. IFM is typically used
with the lowest level of excitation light possible so that the probe can be traced for a relatively long period of time. This usually gives rise to a relatively low signal-to-noise ratio,
which needs to be handled as part of the image processing procedure.

6-26

6.4 Cell Rolling Velocity

27

Recording
device

Microscope

Camera

Time code
generator

Frame
grabber

Image
processing
computer

Monitor

Figure 6.16 Intensified Fluorescence Microscopy system. In this system, the fluorescent signal is
received through the microscope by the camera. The video image is typically time stamped and sent
to the frame grabber. The frame grabber digitizes the signal and sends it to the computer for image
processing. The processed signal is then sent to a monitor for display by the operator and permanently recorded for later playback and analysis.

6.5 Cell Pore Size Determination

Cell pore size is an important parameter for a number of reasons. When a cell secretes
hormones, neurotransmittters, or enzymes, a pore is opened in the cell, which allows
these molecules to be released (Hoffman, 1997). Also, the first indication that two cells
are going to fuse is the formation of a small pore connecting the two cells. The understanding and regulation of these processes are important to the formation and growth of
artificial tissue.
6.5.1 TEM

The transmission electron microscope (TEM) (discussed in Chapter 4) is a useful tool for
determining pore sizes because of its resolution in the 0.3 to 1 nm range. Electron microscopes did not typically view larger objects well so it was difficult to view the cell as a
whole, locate a point of interest, and then zoom in on it. Modern electron microscope
systems, however, allow for this zooming, and the observation of whole tissues while
retaining macromolecular resolution. TEMs work in a vacuum, so living cells cannot be
observed with this type of microscope. Also, most materials in the cell are transparent to
electrons, so they must be stained with dyes. Thus it is possible that the cellular structure
being imaged may have been altered by the dye.

Figure 6.17 A TEM image of skeletal muscle cells.

6.5.2 SEM

The scanning electron microscope (SEM) (discussed in Chapter 4) is also useful in cellular analysis because of its ability to show cell surface detail at resolutions in the 3 to 10
nm range and its ability to show 3-D structure. The 3-D structure is obtained through the
process of Secondary Emission Ion Scanning. As the electron beam of the scanning microscope scans the surface of an object, it can be designed to etch or wear away the outermost atomic layer. The particles are analyzed with each scan of the electron beam. Thus,
the outer layer is analyzed on the first scan, and subsequently lower layers analyzed with
each additional scan. The data from each layer are then analyzed and reconstructed to

6-28

6.5 Cell Pore Size Determination

29

produce a 3-D atomic image of the object. Since electrons are relatively small, the etching is sometimes enhanced by bombarding the surface with ions rather than electrons.

Figure 6.18 A SEM image of stressed liver cells.

6.6 Cell Deformation

Cell deformation involves measuring the mechanical properties of cells and the forces
involved in cell activation. If a force is applied to a cell and the deformation is measured,
the mechanical properties of the cell can be determined. The deformation of cells, in response to some mechanical force outside the cell, is an area of great interest to biomedical researchers. This process is essential in developing mathematical models of the biomechanical aspects of cellular structure, particularly as they apply to the circulatory system. This research involves in vitro biological experiments, mathematical modeling of
cellular mechanical behavior, and computer simulation of cell mechanics. Looking at cell
deformation, researchers are trying to determine the influence of deformation on the
growth of cells, how cells sense deformation at their surface, and how they change their
function and structure in response to this deformation. Typically video microscopy is
used to monitor, record, and measure the amount of deformation in these experiments.
The more difficult engineering task with this parameter is the ability to apply a measurable force to deform the cell.
6.6.1 Micropipet technique

Several methods can be used to deform cells (Hochmuth, 1990). The micropipet has been
an important tool for cell deformation studies. This technique involves aspirating part or
all of a cell into a glass micropipet. A pipet with a diameter of 1 m is capable of producing a force F of 1 pN. The rigidity and other cell deformation properties can be found as a
function of the force on the cell and the displacement of the cell inside the micropipet. By
tracking the displacement of the leading edge of the cell as it moves into the pipet, cell
deformations as small as 0.1 m can be measured.

Cell

Micropipet
Displacement

Figure 6.19 In the micropipet technique, force causes a displacement to determine cell deformation
properties.

6.6.2 Optical trapping

6-30

6.5 Cell Pore Size Determination

31

Another technique used is called optical tweezers or a laser trap. This system develops an
optical gradient generated by a continuous laser. The optical trapping system uses an inverted microscope in differential interference contrast mode and a motorized xy stage
and two associated lasers. The microscope feeds a video camera and the video is recorded
and stored on an optical memory disk recorder. The optical tweezers can be moved
throughout the viewing field and cell deformations as small as 0.1 m can be measured.

6.7 Cell Shear Stress

When a liquid flows through a vessel of uniform diameter, the liquid adopts a parabolic
distribution of velocities from zero at the wall to a maximum at the center. Shear stresses
are generated due to regions of the liquid flowing at different speeds. A different velocity
of the liquid flowing on either side of a cell creates pressure on the cell (Figure 6.20).
This is of particular interest to bioengineers working with the circulatory system. Cell
shear stress is defined as the cellular response to this mechanical effect within the body.
Endothelial cells produce many different kinds of compounds that control vascular tone,
vessel diameter and blood flow rate. The production rate and the release of these compounds are modulated by blood flow and the associated shear stress. Tissue engineers
need to understand the effect of physical parameters, such as pressure and shear stress, on
tissue formation (Lewis, 1995). For example, bone marrow cells cultivated in a petri dish
produce only two of ten types of blood cells. When these cells are cultivated in a bioreactor, a device that simulates some of the physical parameters of the body, all ten types of
red blood cells are produced.
There are two problems in dealing with cell shear stress. One is knowing what is
going on in the cell when the shear stress is applied. The other is what effect the shear
stress has on the cells.
Vessel wall
Blood cells

V1
V2

Blood

Range of
velocities

Figure 6.20 Parabolic distribution of velocities and the shear stress it causes on a blood cell.

6.7.1 Cone and plate

One of the problems with shear stress measurements involving the circulatory system is
that most methods of generating shear stress require relatively large amounts of blood.
However, the cone and plate method solves this problem. With this system, the liquid is
placed between a flat horizontal plate and a cone-shaped surface (Eylers, 1992). The cone
is rotated at a constant speed. This causes the liquid to rotate at speeds that are linearly

6-32

6.7 Cell Shear Stress

33

proportional to their distance from the axis of rotation since the slope of the cone is constant. The shear stress S is the force per unit area of the wall and is defined as
S =

dv
dx

(6.4)

where is the viscosity of the liquid (0.001 Pas for water at 20 C), and dv/dx is the gradient of the velocity between the cone and the plate. Figure 6.21 shows that with this constant velocity gradient, the shear stress is proportional to the viscosity of the liquid. If the
viscosity remains relatively constant, as is the case with blood, the shear stress remains
constant. Changing shear stress values can then be accomplished by changing the speed
(v) of the rotating cone.

Liquid

Cone

dv/dx
Plate (microscopic slide)

v=0

vmax

Figure 6.21 Cone and plate system. The cone is immersed in a liquid and then rotated at a constant
rate. Since the velocity changes linearly with distance, the shear stress is a function of rotation velocity.

6.7.2 Fluorescent ratiometric imaging

Shear stress is reported to induce calcium increase in the cell. Fluorescent ratiometric
imaging (see section 6.13.1) is used to analyze the calcium response with respect to shear
stress. The fluorescent image is an image pair acquired at two different wavelengths and
obtained by a silicon-intensified target camera. After the signals have been sent to an
image processing system and calibrated, the images display the calcium concentration
within the cells.

6.8 Cell Adhesion

Cell adhesion, like cell rolling velocity, is an important parameter in the healing process,
in facilitating the growth of tissue, and in the bodys ability to assimilate newly implanted
biomaterials.
In some cases reduction of adhesion is the aim, in others an increase is desired.
Understanding the mechanisms that start the adhesion process, especially in the bloodstream, are of particular interest. In addition, the strength of adhesive bonds is of prime
interest to the tissue engineer. Most studies of cell adhesion use force measurements obtained from video and fluorescence microscopy imaging.
6.8.1 Laser trap

The optical tweezer, or laser trap, method is used for generating controlled forces in the
piconewton range. It uses a light microscope and the force of radiation pressure from an
infrared laser. Although piconewton sized forces are quite small, they are sufficient to
stop the motion of highly mobile cells or single enzymes. These optical tweezers allow
for measurements of subcellular components including torsional stiffness, flexibility of
the cell cytoskeleton, and forces generated by enzymes.
6.8.2 Interferometry

Interference microscopy is used for examining cell adhesion. It provides information on

the patterns of adhesion, which in turn provides information on the distribution of forces
and how the attached areas of cells move. An interferometer splits a light beam into two
beams, which are then recombined so they can interfere (Chapter 5). The difference in
optical path length between the two beams causes the interference. Changes in optical
path length occur with changes in refractive index (changes in density) of the structure on
which the beams are focused. Therefore, the image shows different interference patterns
in areas with different densities. Interference microscopes are operated in both the reflectance and transmittance modes.
6.8.3 SEM

The SEM uses an intense electron source known as the field emitter, which provides
resolution to diameters less than 1 nm (Morton, 1997). This is used to study the cell adhesion molecule. These molecules are found on the surface of leukocytes and are an important mechanism in the bodys response to injury.

6-34

6.9 Cell Migration

Cell migration is the process by which cells move to a certain area, especially during
growth or in response to an abnormal condition in the body. Cell migration is an area of
great interest for tissue engineering applications such as tissue growth and biomedical
implantation.
6.9.1 4-D image processing system

A four-dimensional (4-D) image processing system is used to create a digital movie from
time-sequences of 3-D cellular tomographic microscope images. First, a single point image is scanned over a 2-D plane. Vertical slices of the 2-D plane are acquired and a 3-D
image is reconstructed. This process must be repeated for each point in time. Then each
of these 3-D images is displayed sequentially and the result is an animation of the movement within the image as a function of time. Typically, this animation is recorded on a
video recorder for ease of repetitive observation and analysis or is assembled into a computer animation file. Although the animation does not improve the quality of the collected
data, it improves the qualitative interpretation of the process.
One problem with this technique is the large amount of storage space and computing power required for the image processing. Cell migration observations typically
take 24 to 96 h to complete (Di Francesca and Zerbarini, 1994). Four fields of view, each
containing 5 to 20 cells, are collected at 15 to 30 min intervals and the position of each
cell is recorded in each frame at every event interval. The data are given to modeling and
analysis software to determine where cells moved and how fast they moved there.

6-35

6.10 Cell Uptake

Cell uptake is the study of the various types and amounts of biological molecules in the
cell. Cell capacitance has been used as an index of secretion in cells. The capacitance of
the cell membrane can be measured using techniques similar to the voltage clamp. Fluorescent microscopy is commonly used since fluorescent probes are now available to
measure pH, sodium, magnesium, potassium, calcium and chloride ions.
6.10.1 Nanovid microscopy

To follow structures within a cell, a probe needs to be sensitive to small amounts of the
structure and only label the structure of interest. Submicron colloidal gold particles coupled to antibodies are useful for this purpose (Geerts et al., 1991). These small particles
can be detected with a light microscope since the particles scatter a small amount of light,
which is detectable using video contrast enhancement microscopy. The particles appear
as black dots on a white background and the technique is called nanovid microscopy. The
biggest advantage of this technique is its ability to label specific molecules within the cell.
Another advantage is the ability for continuous observation without having to worry
about the photobleaching effects seen in fluorescent probes. A final advantage is the
small size of these particles, which allow diffusion throughout the cytoplasm.
6.10.2 Electroporation

Electroporation is a process that is used to temporarily alter cell membranes by inducing

microscopic pores to open in the cell membrane. This allows cell uptake of various biological molecules. This process involves creating an electric field via a voltage pulse.
With the proper pulse shape, the cell completely recovers, the pores reseal, and the cell
continues to grow. An advantage of this technique is that it can be used with certain cell
types that are affected or killed by chemical methods.

6-36

6.11 Cell Protein Secretion

Cell protein secretion involves the movement of proteins across biological membranes
and is not well understood. Epithelial cells from certain organs produce and secrete many
important proteins that are stimulated by specific hormones and neurotransmitters. Also,
cell surface receptor proteins are secreted allowing cells to stick together and form the
extracellular matrix. As such, cellular engineers are focusing on the regulation of the external and internal environments of the cell that promote the secretion of specific proteins.
6.11.1 Atomic force microscopy

Atomic force microscopy (AFM), as discussed in Chapter 4, allows the magnification of

electron microscopy with the ability to work with samples that have not been dehydrated.
This resolution allows for observation of individual protein molecules in action. This
relatively new technology is being used primarily to provide topological mapping of substances but is also being used to detect variations in composition, adhesion, and friction
(Digital Instruments, 1995). For example, AFM permits study of wear debris a few hundred nanometers in size from ultrahigh molecular weight polyethylene (UHMWPE) used
in hip implants.
6.11.2 Fluorescence lifetime imaging

In the fluorescence lifetime imaging microscopy (FLIM) method, image contrast in the
fluorescence microscope is derived from the fluorescence lifetime at each point in the
image. This method is also known as Time-Resolved Fluorescence Microscopy (TRFM)
and is analogous to magnetic resonance imaging (MRI) discussed in Chapter 7. It provides chemical information about intracellular ions since the lifetimes of many probes are
altered by the presence of certain ions and/or pH. The fluorescence lifetimes of the
probes are mostly independent of their concentration and are not normally affected by
photobleaching. The most desirable feature of FLIM, however, is that it does not require
wavelength specific probes. Therefore, FLIM allows quantitative imaging using visiblewavelength illumination. There are several methods involved in this technique (Tian and
Rodgers, 1990). Figure 6.22 shows the time correlated single photon counting method. In
this system, the time difference between the start of an excitation laser light pulses arrival at the sample and the arrival of the fluorescence photon at the detector is measured.
The detector output is recorded in a multichannel analyzer with the intensity at a given
channel being proportional to the number of data points (the number of photons) that
were collected during that time interval.

6-37

6-38

Chapter 6 Cellular Measurements

Microscope

Light
source

Beam
splitter

Photomultiplier
tube

Sample

Start timer
circuit

Stop timer
circuit

Timer

Time-toamplitude
converter

Multichannel
analyzer

Figure 6.22 Time-correlated single photon counting system block diagram. When a light source
provides an excitation pulse, some of the light is deflected to the start timer circuit while the rest
illuminates the sample. A single fluorescent photon is detected by a photomultiplier tube, which
generates a pulse that stops the timer. The time difference is then converted to an amplitude. The
various amplitudes are recorded in the multichannel analyzer and a profile of different time intervals versus number of photons in that interval is displayed.

The other method using this technique simply monitors the fluorescent intensity
over time. This measurement is normally achieved with a fast-response photodetector and
a sampling oscilloscope for data acquisition. The sample is excited and the fluorescent
intensity of the sample is recorded over time.
6.11.3 Fluorescence recovery after photobleaching

Fluorescence recovery after photobleaching (FRAP), also known as fluorescence photobleaching recovery (FPR), is a technique used to look at the mobility of plasma membrane proteins within the cell (Peters and Scholz, 1990) by using the effect of photobleaching in a positive way. First, a probe is introduced that binds to plasma membrane
proteins. Then a laser with a low-power beam is focused on a small circular area on the
plasma membrane. This causes the probe in that area to fluoresce. The power of the laser
is then increased by several orders of magnitude for 0.1 to 1.0 s. This causes the probes in
that area to photolyse (convert to a nonfluorescent product). The laser is turned back to
the original low level and initially the area is dark since all the fluorescent probe material
no longer fluoresces. With time, plasma membrane proteins from neighboring areas diffuse into the circular area and the fluorescence recovers with time as in Figure 6.23.

39

6.11 Cell Protein Secretion

F()
F()

Level of
fluorescence

B
F(+)

Photobleaching

Figure 6.23 A sample graph of fluorescence of proteins during FRAP. (a) Before photobleaching
F(), (b) just after protein photolysed F(+), and (c) long after protein photolysed F().

If not all the proteins are mobile within the membrane, the percentage of mobile proteins
can be calculated using
Rm =

F ( ) F (+ )
F ( ) F (+ )

(6.5)

where Rm is the fraction of mobile proteins, F(), F(+) and F() are the levels of fluorescence (proportional to light intensity) at times immediately before, after, and long after
the area has been photolysed.

6.12 Cell Proliferation

Cell proliferation is the rate at which cells multiply and grow. Many tissue engineering
studies depend on the ability to determine the effect of various factors on cell proliferation. These include chemicals and/or conditions that may influence the rate and/or timing
of cell proliferation. These studies that help in the understanding of the regulation of cell
proliferation are key in the development of new tissue engineering methods and the biomaterials used to support them. This is also an important parameter in developing more
successful therapies for such things as the prevention and treatment of cancer and the
screening of anticancer drugs. Cell proliferation studies are used when measuring growth
factor, drug screening, and the study of toxic, mutagenic, and carcinogenic effects of
chemical compounds on cells (Roche Molecular Biochemicals, 2000). The most basic
method involves using a microscope counting chamber, or hemocytometer, to count the
number of cells in a given area. Density of cells growing on a vessel surface is useful for
tissue engineering applications. Density is measured through the use of an ocular grid
inserted into an inverted phase contrast microscope. Cell density is recorded and then
plotted on a log scale against time in culture. However, more modern techniques are now
being used. There are two general methods of measuring cell proliferation in cell populations including measuring DNA synthesis and measuring metabolic activity.
6.12.1 Radiation monitoring

One method of measuring DNA synthesis involves using a radioactive tracer by adding it
to a cell culture. Over time, some of the radioactive material is retained in the DNA of the
cells. The cells are washed through filters and the radioactivity of the filters containing
the radioactive cells is measured using liquid scintillation counting. Knowing the radiation level of the original tracer in the cell and comparing it to the current radiation level
of the cell provides a ratio that is proportional to the number of times the cell has divided.
6.12.2 Fluorescence microscopy

Fluorescence microscopy techniques are used in a way similar to radiation monitoring.

Fluorescent probes are fed to cells, which are given time to incorporate them into their
DNA (typically 2 to 24 h). Then a measurement of the fluorescence level of collected
cells can be used to determine the level of proliferation of the cell population.
One other method involves using special salts and detects whether cell proliferation is occurring. The salts react with metabolically active (proliferating) cells to form a
red formazan salt, which is easily detected using a colorimetric assay (Chapter 3).

6-40

6.13 Cell Differentiation

Differentiation of a cell is defined as the process where cells become structurally and
functionally specialized during embryonic development. There are 200 different kinds of
cells in our bodies, yet all are developed initially from a single cell. There are many factors that affect cell differentiation, a couple of them being genetics and the environment.
Most research on cell differentiation focuses on understanding the mechanisms involved
in cell differentiation. This is especially important in tissue engineering. Learning what
cell types are involved in tissue regeneration is critical in building tissue. Understanding
the mechanisms of cell differentiation aids in the understanding of what a cell needs to
regenerate a specific part.
Staining techniques are widely used to visualize those components that are otherwise too difficult to see under an ordinary light microscope because of the lack of color
contrast between the object under examination and the background or because of the limited resolving power of the light microscope. In addition, staining techniques are useful in
detecting the presence or absence of certain cell components, thus allowing a relatively
simple scheme of differentiation or identification of microorganisms. The dye used in
staining varies with the cell types.
6.13.1 Fluorescence ratio imaging

Fluorescence ratio imaging is used for measuring changes in concentrations of physiological parameters within cells. This is accomplished by taking advantage of the property
that fluorescent probes have a wavelength shift when they bind with ions in the cell. A
ratio of the intensity of light at the wavelength of emission of the probe itself and the
light intensity at the wavelength of emission of the probe that has bound itself to ions in
the cell is determined. This ratio of fluorescence intensities at the two wavelengths is then
used to calculate the relative concentration of the probe to the concentration of the ion
(Bolsover et al., 1993).
First the background must be subtracted from the two signals.
I 1 = I 1 B1
I 2 = I 2 B 2

and then the ion concentration ratio (R) is calculated using

R=

I 1 S 2
I 2 S 1

(6.6)

where 1 and 2 are the two excitation wavelengths, I1 and I2 are the fluorescence
intensities, B1 and B2 are the background light levels, and S1 and S2 are the shading

6-41

6-42

Chapter 6 Cellular Measurements

corrections that compensate for the small spatial distribution between the two excitation
beams. This method has probes that allow ratio imaging of Ca2+, K+, Mg2+, Cl and pH.
6.13.2 Cell sorter

Flow cytometry systems (Chapter 5) examine cells previously labeled with fluorochromes one at a time and then separate the cells on the basis of their color of fluorescence. Fluorescence activated cell scanning and sorting uses a flow cytometer and a PC
workstation for data processing and analysis. The system uses multiple excitation wavelengths in the visible and UV range. Cells may be separated into a wide variety of tissue
culture containers. The PC can be used for statistical purposes providing histograms of
the various distributions of cells.

6.14 Cell Signaling and Regulation

Cell signaling and regulation involves the complex processes that occur within cells to
maintain life (regulation) and trigger new events to occur (signaling). For the biomaterials engineer, the dream is to manufacture material that could (1) sense its environment,
(2) integrate the sensed information and finally (3) take some action based on it. To do
this, engineers must study the processes of signaling and regulation within the cell and
develop an understanding of how and why they work as they do.
A number of techniques are used for research in this area. NMR provides noninvasive images of cellular structure and chemical composition. This technique is discussed
in detail in Chapter 7. Microscope Laser Light Scattering Spectroscopy (MLLSS) is used
for motion studies of intercellular macromolecules during diffusion or flow. This method
directs laser light into a sample and subsequently detects the scattered light. Fluorescent
probes are used to mark specific membrane proteins, which can in turn be microinjected
into cells. Their paths through the cell are monitored by computer-aided video systems.
With these systems, both quantitative and qualitative observation of membrane flow in
live cells can be monitored. In addition, time lapse photography coupled with phase contrast microscopy allows visualization of the cell division process. However, two of the
most common methods are described below.
6.14.1 FISH

Fluorescence In Situ Hybridization (FISH) is a technique for detecting specific DNA

sequences in chromosomes, whole cells, and tissue slices. In situ hybridization is the
process of attaching a small fragment of DNA to a specific target strand of DNA or RNA
in a preserved cell. Probes are used that only attach to a specific genetic sequence in that
cell. Then the point of attachment along the chromosome is identified (Figure 6.24). This
method is useful if the identity of a particular gene is known but the chromosome on
which it is located is not. FISH identifies the chromosome and the exact spatial location
of the gene by exciting the probe and then determining its spatial location. Another application is in the study of embryo development and the mechanisms involved in the sequence of gene selection, activation, and location. Using a CLSM and multiple different
colored probes, analysis of fifteen or more probes simultaneously is possible. To do this,
the spectrum of the fluorescence emission must be determined since it may contain light
that is a mixture of several different colors. This is accomplished with a technique called
Fourier spectroscopy (Soenksen et al., 1996). An interferometer splits the fluorescent
light emitted by the sample, then the beams are recombined to interfere with one another
(Chapter 4). By changing the length of the optical path difference between beams, an
intensity versus optical path difference is obtained. This result is sent to a computer
where a spectral profile of light intensities for all wavelengths of interest is calculated
using a Fourier transform technique. Since the probes used typically have a relatively

6-43

6-44

Chapter 6 Cellular Measurements

large span between emission wavelength, the intensity peaks where each probe is fluorescing is readily apparent from the spectral image.
The 3-D organization of specific chromosomes within cells and tissues is also
being examined using FISH in combination with laser scanning confocal microscopy and
computer assisted 3-D image reconstruction algorithms.

Fluorescent probe

Chromosome target

Figure 6.24 (a) The DNA strand is denatured with heat and put in the same culture as the fluorescent probe. (b) The fluorescent probe binds to the target gene.

Figure 6.25 Human chromosomes probed and counterstained to produce a FISH image. The lighter
area within each chromosome show the fluorescent probe attached to the target gene. (from Fluorescence Detection: Anti-Dye Antibodies. Molecular Probes Product Literature. [Online] Available:
www.probes.com/lit/fish/part4.html)

6.14.2 Radioactive labeling

A process known as pulse labeling is used to determine the site of molecular synthesis in
cells. A radioactive tracer is added to the cellular environment for a short period of time.
The radioactive media are then washed or diluted away. The cells are then fixed and
autoradiography is used to locate the sites of newly synthesized DNA. The site of utiliza-

6.14 Cell Signaling and Regulation

45

tion of a particular molecule is determined by pulse-chase labeling: cells are again exposed to a radioactive tracer for a short period of time and then washed away. But this
time, the cells are allowed to grow for a period of time. When these cells are examined,
the radioactivity will have moved to the site of utilization in the cell.

6.15 References
Bolsover, S. R., Silver, R. A. and Whitaker, M. 1993. Ratio imaging measurements of
intracellular calcium. In D. Shotton (ed.) Electronic Light Microscopy, Techniques in
Modern Biomedical Microscopy. New York: Wiley-Liss.
Eylers, J. P. 1992. Liquids. In J. F. Vincent (ed.) Biomechanics Materials. Oxford: Oxford University Press.
Geerts, H., De Brabender, M., Nuydens, R. and Nuyens, R. 1991. The dynamic study of
cell surface organization by nanoparticle video microscopy. In R. J. Cherry (ed.)
New Techniques of Optical Microscopy and Microspectrometry. Boca Raton, FL:
CRC Press.
Heidecamp, W. H. 1996. Cell Biology Laboratory Manual. [Online] Available:
www.gac.edu/cgi-bin/user/~cellab/phpl?chpts/chpt1/intro1.html
Hochmuth, R. M. 1990. Cell biomechanics: A brief overview. Trans. ASME, J. Biomech.
Eng., 112 (3): 233-4.
Hoffman, M. 1997. Cellular Secretion: Its in the Pits. Am. Sci., [Online] Available:
www.sigmaxi.org/amsci/Issues/Sciobs97/Sciobs97-03pits.html
Haugland, R. P. 1997. Handbook of Fluorescent Probes and Research Chemicals.
Eugene, OR: Molecular Probes Inc.
LOCI 2000. Multiple Photon Excitation Fluorescence Microscopy. [Online] Available:
www.loci.wisc.edu/multiphoton/mp.html
Lewis, R. 1995. Tissue engineering now coming into its own as a scientific field. The
Scientist. 9: 15.
Molecular Probes. 2000. Product Literature. Fluorescence Detection: Anti-Dye Antibodies. [Online] www.probes.com/lit/fish/part4.html
Paesler, M. A. and Moyer, P. J. 1996. Near Field Optics: Theory, Instrumentation and
Applications. New York: John Wiley & Sons.
Pawley, J. B. 1990. Handbook of Biological Confocal Microscopy. New York: Plenum
Press.
Peters, R. and Scholz, M. 1991. Fluorescence photobleaching techniques. In R. J. Cherry
(ed.) New Techniques of Optical Microscopy and Microspectrometry. Boca Raton,
FL: CRC Press.
Photometrics 2000. CCD Grading. [Online] Available:
www.photomet.com/library_enc_grading.shtml
Richardson, M. 1997. Three Dimensional Deconvolution of Microscope Data. Vaytek,
Inc. [Online] Available: www.vaytek.com/technical.html
Roche Molecular Biochemicals 2000. Guide to Cell Proliferation and Apoptosis Methods.
[Online] Available: biochem.roche.com/prod_inf/manuals/cell_man/cell_toc.html
Soenksen, D. G., Garini, Y. and Bar-Am, I. 1996. Multicolor FISH using a novel spectral
bio-imaging system. In Asakura, T., Farkas, D. L., Leif, R. C., Priezzhev, A. V.,
Tromberg, B. J. (eds.) Optical Diagnostics of Living Cells and Biofluids. Proc. SPIE,
2678.

6-46

6.15 References

47

Shotton, D. (ed.) 1993. Electronic Light Microscopy, Techniques in Modern Biomedical

Microscopy. New York: Wiley-Liss.
Shotten, D. M. 1991. Video and opto-digital imaging microscopy. In R. J. Cherry (ed.)
New Techniques of Optical Microscopy and Microspectrometry. Boca Raton, FL:
CRC Press.
Stine, B. 1995. High Resolution Imaging of Biological Samples by Scanning Probe Microscopy. Application Notes, Digital Instruments. [Online] Available:
www.di.com/AppNotes/ANMain.html
Tian, R. and Rodgers, M. A. J. 1991. Time-resolved fluorescence microscopy. In R. J.
Cherry (ed.) New Techniques of Optical Microscopy and Microspectrometry. Boca
Raton, FL: CRC Press.
Wright, S. 1993. Introduction to confocal microscopy and three-dimensional reconstruction. In B. Matsumoto (ed.) Methods in Cell Biology. San Diego: Academic Press.

6.16 Problems
6.1
6.2
6.3
6.4
6.5

6.6
6.7
6.8
6.9
6.10

6.11
6.12
6.13

6.14
6.15
6.16
6.17
6.18
6.19
6.20

List the size of an atom, molecule, virus, cell.

The object length of a microscope is 10 mm. Calculate the focal length for a
magnification of 20.
Find the resolution of a microscope with an index of refraction of 1.2 and a cone
angle of 80 that is operated at a wavelength of 770 nm.
Explain the meaning of the terms brightfield, darkfield, phase contrast and differential interference contrast in the field of microscopy.
Sketch a phase contrast microscope. Show the light paths and use these to explain how the phase contrast microscope achieves its improvement over a simpler microscope. Explain how the phase contrast microscope is changed to
achieve a dark field.
Explain how a two-dimensional image of incident light is obtained in a solidstate camera.
Describe the term fluorescent probe as it applies to cell microscopy.
Explain the reason for using and the basic principles of operation of fluorescence microscopy.
Describe the effect of photobleaching in fluorescence microscopy.
The radiation counting system shown in Figure 6.10 has a voltage supply of 100
V. Sketch waveforms of voltage versus time for (1) left of the resistor, (2) right
of the resistor, (3) right of the capacitor, when ionization by a radiation particle
causes a voltage pulse. Name the electric circuit component that converts the
analog voltage at the right of the capacitor to a digital signal suitable for input to
a pulse counter.
A radiation detection system records 350 counts in 20 min. The detector has an
efficiency of 0.93. Determine the actual number of decays per minute.
Explain the reason for using and the basic principles of operation of a confocal
laser scanning microscope.
For a confocal laser scanning microscope, for fixed optics, explain how scanning is accomplished in the x, y, and z directions. For nonfixed optics that can
deflect the scanning beam, sketch the location of the scanning mechanism. Explain how scanning is accomplished in the x, y, and z directions.
Explain the reason for using and the basic principles of operation of a twophoton excitation microscope.
Explain the reason for using and the basic principles of operation of deconvolution image processing.
Explain the reason for and the basic principles of measuring cell orientation.
Explain the reason for and the basic principles of measuring cell rolling velocity.
Explain the reason for and the basic principles of measuring cell pore size determination.
Explain the reason for and the basic principles of measuring cell deformation.
Explain the reason for and the basic principles of measuring cell shear stress.

6-48

6.15 References

6.21
6.22
6.23
6.24
6.25
6.26
6.27
6.28
6.29

49

Calculate the shear stress for water at 20 C in a cone and plate viscometer with
a diameter of 10 cm, a separation gap of 1 mm at the circumference, and rotation
at 1 revolution per second. Give units.
Explain the reason for and the basic principles of measuring cell adhesion.
Explain the reason for and the basic principles of measuring cell migration.
Explain the reason for and the basic principles of measuring cell uptake.
Explain the reason for and the basic principles of measuring cell protein secretion.
Using fluorescence recovery after photobleaching, calculate the fraction of mobile protein where F() = 4, F(+) = 1, F() = 2.
Explain the reason for and the basic principles of measuring cell proliferation.
Explain the reason for and the basic principles of measuring cell differentiation.
Explain the reason for and the basic principles of measuring cell signaling and
regulation.

7
Nervous System
Jang-Zern Tsai

The nervous system is defined as all cells, tissues, and organs that regulate the bodys
response to both internal and external stimuli (Costello, 1994). There are many
neurological diseases and disorders that can be assessed by the use of medical equipment.
Biomedical engineers need to understand how the nervous system works in order to
design this equipment.

7.1 Cell Potential

The nervous system is comprised of neuron cells. Neurons are the conducting elements of
the nervous system and are responsible for transferring information across the body. Only
these and muscle cells are able to generate potentials and therefore are called excitable
cells. Neurons contain special ion channels that allow the cell to change its membrane
potential in response to stimuli the cell receives (Campbell et al., 1999).
7.1.1 Resting potential
All cells in the body have a cell membrane surrounding them. Across this membrane
there is an electric charge referred to as the resting potential. This electric impulse is
generated by differential ion permeability of the membrane. In the cells, potassium (K+)
channels allow diffusion of K+ ions out of the cell while sodium (Na+) ions diffuse into
the cell. This Na+K+ pump, which requires ATP to operate, pumps two K+ ions into the
interior of the cell for every three Na+ ions pumped out. K+ and Na+ ions are
continuously diffusing across the membrane from where they were just pumped, but at a
slower rate. Since there are more K+ ions inside the cell than outside, a potential exists.
The magnitude of the equilibrium membrane resting potential is given, at body
temperature, by the Nernst equation

Nervous System
Vm = (RT/nF)ln[K+]e/[K+]I
Vm = 0.0615 log10[K+]e/[K+]i

(7.1)

At 37 C (body temperature), where Vm is the transmembrane potential in millivolts, R is

the universal gas constant, T is absolute temperature in K, n is the valence of the ion, F is
the Faraday constant, [K+]e is the extracellular potassium concentration, and [K+]i is the
intracellular potassium concentration. Typically, [K+]e = 4 mmol/L, [K+]i = 155 mmol/L,
which yields Vm = 97.7 mV. However, other ions such as Na+ and Cl have a lesser
effect on the resting potential because their permeabilities are smaller, so the actual Vm
= 85.3 mV (Clark, 1998). The typical resting potential for a neuron is 70 mV
(Campbell et al., 1999).
7.1.2 Action potential
Some cells, such as skin cells, are not excitable. Other cells, such as nerve and muscle
cells, are excitable. When a stimulating electric field acts on an excitable cell, the Na+
permeability increases, Na+ enters the cell interior, and the entering positive charge
depolarizes (reduces to approximately zero) the transmembrane potential. Later the K+
permeability increases and K+ ions flow out to counter this effect. The Na+ gates close
followed by the K+ gates. Finally, the resting potential is regenerated. See Figure 7.2.
The action potential lasts about 1 ms in nerves and about 100 ms in cardiac muscle. It
propagates in nerves at about 60 m/s and carries sensation from the periphery toward the
brain via sensory nerves. Through motor nerves, the brain commands muscles to contract.
Figure 7.1 shows an invertebrate nerve axon under stimulation of an electric potential
high enough to excite the depolarization of the axon (Stein, 1980). We can calculate the
action potential propagation velocity v = d/t, where d = distance and t = time.

7.1 Cell Potential

Indifferent
electrode

Electronic
stimulator

Micropipet electrode

1m

Stimulus

0V

Action potential

20 ms

Time

Resting potential
70 mV

Figure 7.1 The action potential of the nerve axon in response to electric stimulus. The
depolarization process first occurs at the stimulation site. The action potential then travels
downstream to the recording site where a penetrating micropipet is used to pick up the intracellular
potential with respect to an indifferent reference potential.

Nervous System
Na+

Na+

++++

++

+ + + Outside cell
Plasma membrane

Inside cell

++++

++

+++

K+
1

K+
3

Resting phase
Na+

Repolarizing phase
Na+

++++

++

+++

++++

++

+++

K+
Depolarizing phase

Membrane potential
(mV)

K+

+50
0

Undershoot phase

3
4

50

100

Figure 7.2 The role of voltage-gated ion channels in the action potential. The circled numbers on
the action potential correspond to the four diagrams of voltage-gated sodium and potassium
channels in a neuron's plasma membrane (Campbell et al., 1999).

+7.2 Brain, EEG and Evoked Potentials

7.2.1 Anatomy and function of the brain
The brainstem is a short extension of the spinal cord and serves three major functions: a
connecting link among the cerebral cortex, the spinal cord, and the cerebellum; a center
of integration for several visceral functions; and an integration center for various motor
reflexes.
The cerebellum is a coordinator in the voluntary (somatic) muscle system and
acts in conjunction with the brainstem and cerebral cortex to maintain balance and
provide harmonious muscle movements.
Figure 7.3 shows the cerebrum, which occupies a special dominant position in
the central nervous system. Within the cerebrum are the conscious functions of the
nervous system.

Figure 7.3 The cerebrum contains the frontal, parietal, temporal and occipital lobes. From A. B.
McNaught and R. Callander, Illustrated Physiology, 3rd ed., 1975. Edinburgh: Churchill
Livingstone.

7.2.2 Electroencephalography (EEG)

Nervous System

The electroencephalogram (EEG) provides information about the health and function of
the brain by detecting electric impulses in the brain. EEG can help diagnose conditions
such as epilepsy, brain tumors, brain injury, cerebral palsy, stroke, liver, kidney disease,
or brain death and helps physicians find the cause of problems such as headaches,
weakness, blackouts or dizziness (WFUBMC, 2000).
By removing a portion of the skull, it is possible to insert microelectrodes into
the brain and record action potentials from single cells. Because this procedure is so
invasive, the vast majority of clinical studies are made from electrodes glued onto
standard locations on the scalp as shown in Figure 7.4. These electrodes average the
action potentials from large numbers of cells and therefore do not provide action
potentials from single cells. Some localization is possible by using the monopolar
connection, which is a recording from a single electrode referenced to the average of all
other electrodes. Some localization is also possible by using the bipolar connection,
which is a recording between successive pairs of adjacent electrodes. The EEG is
typically 100 V in amplitude with a frequency response of 0.5 to 80 Hz. Sixteen
channels are usually recorded simultaneously (Hughes, 1994).

Figure 7.4 The 1020 electrode system for measuring the EEG. From H. H. Jasper, The Ten
Twenty Electrode System of the International Federation in Electroencephalography and Clinical
Neurophysiology, EEG Journal, 10 (Appendix), 371375.

The EEG is typically recorded with the subject awake and resting. Figure 7.5(b)
shows that when the eyes are closed, alpha waves with a frequency of about 10 Hz appear.
As the subject becomes drowsy and falls asleep, the frequency decreases and the

7.2 Brain, EEG and Evoked Potentials

amplitude increases. During rapid-eye-movement (REM) sleep, frequency increases and

amplitude decreases. Alpha waves correspond to when the subject is awake, theta waves
indicate sleep, delta waves indicate sleeping well, and finally beta waves indicate REM
sleep (Clark, 1998). Each of the four EEG waveforms can be see in Figure 7.5(a).
In people with epilepsy, brain cells create abnormal electricity that causes
seizures (AAFP, 2000). Grand mal epilepsy spreads throughout the brain, causes
convulsions, and is diagnosed by large amplitude recording from all parts of the scalp.
Petit mal epilepsy is identified by a 3 Hz spike-and-dome pattern and may cause an
interruption in consciousness. Psychomotor epilepsy is identified by 2 to 4 Hz waves with
superimposed 14 Hz waves and causes amnesia or unwanted motor action. Figure 7.5(c)
shows various epilepsy waveforms, though other abnormal waveforms also exist that help
to diagnose other disease. An EEG with no response for 24 h is an indication of brain
death.

(c)

Nervous System

Figure 7.5 (a) Four types of EEG waves. (b) When the eyes are opened, alpha waves disappear. (c)
Different types of epilepsy yield abnormal waveforms. From A. C. Guyton, Structure and Function
of the Nervous System, 2nd. Ed., Philadephia: W. B. Saunders, 1972.

7.2.3 Evoked potentials

An evoked potential is the electric response signal of the brain to a specific stimulus. The
response is usually of low amplitude (<5 V), especially in relation to the background
EEG signal (50 to 100 V), so that stimulus-related cerebral evoked potentials usually
cannot be identified in routine EEG recordings. With the use of the signal averaging
technique, the background EEG and artifacts can be minimized. For example, a stimulus
occurs N = 100 times and each time the response is recorded. All 100 responses at 0 ms
are added. All 100 responses at 1 ms are added, etc. The signal averaged output
waveform is a plot of the added outputs at 0, 1, 2, etc. ms. Because these responses are
synchronized to the stimulus, they add linearly and are N times larger. The interfering
noise potentials from the EEG are random and have an amplitude equal to N . Thus
the signal-to-noise ratio (SNR) increases by a factor of N
N = N (Isley et al.,
1993).
Studies of evoked potential are useful in evaluating the functional integrity of
sensory pathways. They permit the lesions of these pathways, which may not have any
visible symptoms, to be detected and localized. They may be helpful in establishing or
supporting a neurological diagnosis, following the course of certain disorders, and
assessing sensory function (such as vision or hearing) when normal behavioral or
subjective testing is not possible. The electric potentials associated with cognitive
processes can also be recorded through the scalp and are helpful in the evaluation of
patients with suspected dementia. Evoked potentials provide objectivity, because
voluntary patient response is not needed. Clinical abnormalities can be detected by an
increase in latency (the delay between the stimulus and the wave response).
Visual evoked potentials
During visual-evoked potential (VEP) testing, the subject stares at a central spot on a
reversing checkerboard pattern 1 m away. The pattern consists of a counterphase
alternation of dark and light squares without any change in total luminance. The reversing
rate is about twice per second. The evoked potential generated in the occipital cortex is
detected by an electrode placed on the scalp in the midoccipital position with reference to
either the linked mastoids or a midfrontal electrode. A recorded VEP is obtained by
amplifying, synchronizing, and averaging the responses to 100 such reversals. The
procedure is usually done twice to ensure reproducibility and the two eyes are tested
separately. VEP evaluates the visual nervous system from the eyes to the brain. It may
help diagnose certain visual or brain disorders (Gevins and Aminoff, 1988) (WFUBMC,
2000).

7.2 Brain, EEG and Evoked Potentials

Brainstem auditory evoked potentials

During brainstem auditory evoked potential (BAEP) testing, rarefaction or condensation
clicks at about 10 Hz stimulate one ear of the subject and white noise masks the
contralateral ear. Click intensity is usually 60 to 70 dB above the patients hearing
threshold. The evoked responses are smaller than 1 V measured from the vertex with
reference to the lobe of the stimulated ear. BAEPs within 10 ms of the stimulus are
filtered, amplified, and averaged for satisfactory recording. At least two separate results
are recorded for each ear to ensure reproducibility. BAEP examination evaluates the
nerve pathways from the ears to the brain. It may help uncover the cause of hearing and
balance problems as well as other nervous system disorders (Gevins and Aminoff, 1988)
(WFUBMC, 2000).
Somatosensory evoked potentials
During somatosensory evoked potential (SSEP) testing, peripheral nerves are stimulated
with a small electric current through bipolar electrodes on the skin of the subjects arms
or legs. Electrodes placed on the subjects scalp and along the spinal cord can detect a
series of positive and negative potentials following the peripheral stimulation. The
intensity of the stimulus is generally very small. About 1000 responses are averaged to
reduce the interference from endogenous cardiac, motor, and EEG activities and
exogenous electric noise. The examination is repeated at least once to ensure
reproducibility of the results. SSEP assesses the neurological pathways from the nerves in
the arms or legs through the spinal cord to the brain. It can also be used to monitor coma
patients, monitor patients during surgical procedures, and test hearing in infants and
others whose hearing cannot be tested in the standard way (Gevins and Aminoff, 1988)
(WFUBMC, 2000).

7.3 Brain Imaging: X ray

In order to diagnose disease in the brain, we may take a picture using X rays. X rays are
very short wavelength electromagnetic waves produced when fast-moving electrons
collide with substances in their path. They are similar to light rays, except X-ray
wavelengths are only 1/10000 the length of visible light rays. Because of this short
wavelength, X rays can penetrate very dense substances to produce images or shadows
that can then be recorded on photographic film. Radiography is diagnostically useful
because differences in density between various body structures produce images of
varying light and dark intensity on the X-ray film, much like the negative print of a
photograph. Dense structures, such as the skull appear white, whereas air-filled areas,
such as the lung are black.
Figure 7.6 shows an X-ray system. Electrons emitted by the heated filament are
accelerated to the tungsten anode at +100 kV, which emits X rays. An aluminum filter
stops low-energy X rays, because they would not pass through the body to form an image.

10

Nervous System

The X rays travel in all directions, but to prevent patient and operator harm, are shielded
by a collimator so only those used in the image proceed. Secondary radiation could fog
the film, but is stopped by a grid shaped like Venetian blinds. Phosphor screens emit
many light photons for each X-ray photon, thus assisting in darkening the photographic
film. To lower X-ray dose, an image intensifier may be used. The X-ray photon strikes a
phosphor layer in a vacuum tube. Many light photons stimulate a photocathode to emit
many electrons. These are accelerated to strike an output phosphor screen at +25 kV,
yielding a good image with low X-ray dose (Siedband, 1998).

Figure 7.6 The 100 kV X-ray tube generates X rays that form a shadow of the body to expose the
film. Unlike a camera, there are no lenses.

7.4 Brain Imaging: CT

Figure 7.7 shows a computed tomography (CT) X-ray system. It measures X-ray
attenuation of many parallel paths in one plane. Then it repeats this at many angles. The
information is processed by computer to yield a 2-dimensional image of a slice through
the body. Many slices can be used to create 3-dimensional images. Figure 7.8 shows how
the image may be digitally reconstructed by backprojection, in which the attenuation
along each path is uniformly assigned along the path. Thus if there is high attenuation at
one picture element (pixel), the assignment from many angles increases that pixel more
than other pixels. Iterative methods guess at the image, measure the resulting attenuation,
correct to that measured by X ray, then back project the revised information. Water is
used as the benchmark substance as it has a CT value of zero. Analytic methods use
spatial Fourier analysis or convolution techniques. CT can produce cross-sectional
images (slices) of anatomic structures without superimposing tissues on each other.
Improved CT scanners use multiple simultaneous measurements and X-ray tubes that
rotate in 2 s. CT yields images of soft tissue within the brain not possible with
conventional X ray, which is largely blocked by the dense skull. The patient is positioned
in the center of a donut hole, surrounded by the CT scanner. CT is used in the differential
diagnosis of intracranial neoplasms (brain tumors), cerebral infarctions (lack of blood

7.4 Brain Imaging: CT

11

supply to the brain), ventricular displacement or enlargement, cortical atrophy (decrease

in size of the brain), cerebral aneurysms (enlarged sections of vessels in the brain),
intracranial hemorrhage and hematoma (bleeding and subsequent clotting in the brain),
and arteriovenous malformation (AVM-malformed blood vessels). However there do
exist limitations of CT as with all imaging devices. The spatial and contrast resolutions
limit CT in its ability to distinguish normal from pathogenic tissue. For, example, if a
tissue (i.e. tumor) in the body has attenuation similar to that of water, it will not be
detected.

Figure 7.7 In computed tomography, a thin X-ray beam measures the attenuation along a single
path. The system is indexed to obtain attenuation in many parallel paths at one angle and then
repeated at many angles. The information is processed to yield attenuation at every element in the
2-dimensional slice. From R. A. Brooks and G. Di Chiro, Theory of image reconstruction in
computed tomography. Radiology, 117: 561572, 1975.

12

Nervous System

1/3 + 1/3
=
2/3

1/3 + 1
=
4/3

1/3 + 1/3
=
2/3

1 + 1/3
=
4/3

1+1
=
2

1 + 1/3
=
4/3

1/3 + 1/3
=
2/3

1/3 + 1
=
4/3

1/3 + 1/3
=
2/3

Figure 7.8 Creating images using backprojection. Each row and column signifies a certain path
angle. The values of 3 in the center row and column are uniformly backprojected with a value of 1
in each cell. These sum to the maximum value of 2, thus indicating an object in that cell. The
values of 1 are uniformly backprojected with a value of 1/3 in each cell. A sum of these two
numbers that is significantly greater relative to the others indicates an object.

7.5 Brain Imaging: MRI

Magnetic resonance imaging (MRI) provides two-dimensional images that are slices of
1H isotope concentration in three-dimensional objects. The patient is placed in a very
strong z-axis magnetic field of about 2 T. The charged 1H electrons spin and precess like
a spinning top at the Larmor frequency of 42.57 MHz/T. The magnetic field is perturbed
to produce a small magnetic gradient along the y-axis. Figure 7.9 shows that there is only
one slice at a particular Larmour frequency. Surrounding RF coils are pulsed at this
frequency, which excites particles in the slice. The magnetic field is quickly perturbed
along the x-axis. An RF receiver measures the relaxation along a single line along the zaxis orthogonal to both gradient fields. The z-axis magnetic field can be perturbed to
produce additional scan lines. The system is similar to CT and the signals are processed
in the same way. The strengths of MRI include: sensitivity to flow (i.e. blood); electronic
adjustment of imaging plane (opposed to physical adjustment in CT); no ionizing
radiation (only magnetism and radio waves); and high soft tissue resolution (see Figure
7.10). However some of the weaknesses are contraindications due to implanted medical
devices and metal plates in patients, slow image acquisitions and high cost.

7.5 Brain Imaging: MRI

13

One slice
in x axis

One slice
in y axis
Single line
y

Figure 7.9 In MRI, RF coils excite one slice in the y-axis. An RF receiver measures from one slice
in the x-axis.

(a)

(b)

(c)

(d)

Figure 7.10 MRI images of the brain.(a) A sagittal view of a normal brain. (b) A sagittal view of a
patient with an arterio-venous malformation (AVM) which is an explosive growth of the capillary
bed. (c)(d) Brain images from the patient with the AVM in which the visual cortex of the brain was
activated using flashing lights while rapid MR imaging was performed. The regions of the brain
that were "activated" by the visual stimulus are displayed in white on the images. Images are from
University of Wisconsin-Madison fMRI (functional magnetic resonance imaging) laboratory.

7.6 Brain Imaging: Nuclear Imaging

14

Nervous System

A radionuclide is an atom with an unstable nucleus. In an attempt to reach stability, the

radionuclide emits one or more types of radiation, the most common examples being
alpha particles, beta particles, and gamma electromagnetic radiation. In general, gamma
rays are used for imaging organ systems and provide an indication of how well an organ
system functions. Computerized radiation detection equipment, particularly scintillation
detectors, detects gamma rays by giving off a light flash (scintillation). The imaging
device provides an image of the organ under study and provides information on its size,
shape, position, and functional activity. It is used mainly to allow visualization of organs
and regions within organs that cannot be seen on a simple X-ray film. Space-occupying
lesions, especially tumors, stand out particularly well (Early and Sodee, 1995).
A radiopharmaceutical is administered orally or intravenously to the patient. A
sufficient time interval is allowed for the radioactive material to follow its specific
metabolic pathway in the body and to concentrate in the specific tissue to be studied. An
imaging device records the position and concentration of the radiation that emerged from
the radionuclide. In almost all instances, radionuclide imaging exposes the patients to less
radiation than would be received undergoing a similar procedure with diagnostic X-ray
studies. With a nuclear medicine scan, metastatic disease to the bone can be found 6
months to a year before it can be detected with the usual bone radiograph. Nuclear
imaging images function rather than anatomy. A nuclear brain scan is used for the
diagnosis of pathologic abnormalities such as tumors, cerebrovascular aneurysms, and
hematomas.
Gamma camera
Figure 7.11 shows a gamma camera. Gamma rays from the organ travel in all directions,
but the collimator tubes ensure that only radiation from a small region reaches the NaI
detector. Light is detected by more than one photomultiplier tube, but the relative
amounts received improve spatial resolution. An electronic pulse height analyzer selects
events that have the proper gamma-ray energy. The resulting image has spatial resolution
of about 1% of the image dimension (Mettler and Guiberteau, 1991).

7.6 Brain Imaging: Nuclear Imaging

15

Figure 7.11 In a gamma camera system, radioisotopes emit gamma rays, which are collimated,
strike a NaI crystal, which emits light measured by photomultiplier tubes.

7.7 Brain Imaging: Single-Photon Emission Computed Tomography

(SPECT)
In single-photon emission computed tomography (SPECT), a scintillation assembly
similar to a gamma camera is rotated around the patient. The gamma rays are collected
from the patient in a manner similar to CT, but several slices can be obtained at the same
time. The resulting multiple slices show depth activity in the volume of interest. It is
possible to see anomalies not observable with conventional X rays or gamma camera
(Van Heertum and Tikofsky, 1995).

7.8 Brain Imaging: Positron Emission Tomography (PET)

Some isotopes produce positrons that react with electrons to emit two photons at 511 keV
in opposite directions. Figure 7.12(a) shows the two detectors on opposite sides of the
patient. The detectors determine if the two scintillation effects are coincident and have
energy levels close to 511 keV. Additional pairs of detectors permit faster operation.
Image reconstruction is similar to that of CT. An advantage of PET is that all of the most
common radioisotopes used, 15O, 13N, 11C and 18F, can be compounded as metabolites.

16

Nervous System

For example, CO can be made with 11C. If a portion of the brain is active, increased
blood flow carries the isotope to it, where it shows up on the image. Abnormal
functioning, tumors, seizures and other anomalies may also be mapped this way. For
example, measurement of glucose-fluorodeoxyglucose (FDG) metabolism is used to
determine tumor growth. Because small amounts of FDG can be visualized, early tumor
detection is possible before structural changes occur and would be detected by MRI or
CT.

Figure 7.12 In the PET camera, (a) the paired and (b) the hexagonal ring cameras rotate around the
patient. (c) The circular ring does not rotate but may move slightly to fill in the gaps between the
detectors.

7.9 Brain Imaging: Biomagnetism

Action potentials within the brain cause ionic currents to flow through cell membranes
and between cells. Current flow creates magnetic fields. Magnetic shielding around the
subject can remove interference from the earths magnetic field of 50 T. A gradiometer
(two coils series opposed) can remove the interfering urban noise magnetic fields of 10 to
100 nT from distant sources, yet detect the nearby sources from the brain. The
Superconducting Quantum Interference Device (SQUID) magnetometer has a sensitivity
of 0.01 pT and can detect the magnetoencephalogram (MEG) of the alpha wave of about
0.1 pT (Malmivuo and Plonsey, 1995) (NATO, 1983).
Advantages of the MEG are that the magnetic field is unaffected by the high
resistivity of the skull. In fact, no electrodes are required on the scalp. Field distributions
of the brain are measured by arrays of 300 or more detectors surrounding the brain (Hari
and Lounasmaa, 2000). The combination of functional information, deduced from MEG,
with structural information obtained from MRI has become almost routine in most MEG
research groups.

7.10 Eye, ERG, EOG and Visual Field

7.10.1 Electroretinogram (ERG)

7.10 Eye, ERG, EOG and Visual Field

17

When the retina is stimulated with light, an action potential is superimposed on the
resting potential. Measuring the electrical response of the retina to flashes of light detects
inherited retinal degenerations, congenital (existing from birth) retinal dystrophies, and
toxic retinopathies from drugs or chemicals. The electrical response of the retina to a
reversing checkerboard stimulus can help distinguish between retinal and optic nerve
dysfunction to diagnose macular disease (Vaughan and Asbury, 1995). Macular disease is
the leading cause of blindness in elderly Americans today and it is also linked to smoking.
This disease affects the central area of the retina, causing gradual loss of central vision.
The degeneration is caused by the partial breakdown of the insulating layer between the
retina and the choroids (the layer of blood vessels behind the retina) (Yahoo! Health,
2000).
The retina has electrical dipole properties that result in the corneal surface being
electropositive with respect to the outer scleral surface. Figure 7.13 shows that the record
is made by placing an active electrode on the cornea, usually one embedded in a corneal
contact lens, with saline solution bridging the gap between the electrode and the cornea,
and placing an indifferent electrode on the forehead. An alternative to the contact lens
electrode would be to use electrodes that hook over or rest on the lower eyelid. These
electrodes require a large number of responses, but are usually a better option for children
than the contact lens electrodes. The entire retina is stimulated with light and the small
voltage is amplified and recorded. The retina is stimulated with light after either dark
(scotopic) or light (photopic) adaptation (Newell, 1996).

Figure 7.13 The transparent contact lens contains one electrode. The reference electrode is placed
on the right temple.

18

Nervous System

Flash ERG
In flash ERG, the inner and outer nuclear layers of the retina produce a and b waves,
respectively (Figure 7.14). Some of the cells that make up the two layers of the retina are
the photoreceptors, horizontal cells, bipolar cells, and Muller cells. All of the cells in the
two layers give a graduated hyperpolarizing or depolarizing response to light, as
compared with the action potentials produced by the retinal ganglion cells. The a wave of
the ERG is produced by the hyperpolarizing photoreceptors. This wave has separate rod
and cone components. The b wave is generated from the activity of depolarizing bipolar
cells. The a wave represents photorecptorial activity and the b wave represents
postphotoreceptorial activity in the second-order retinal neuronal network, but the
sources are still unclear.
To differentiate between rod and cone activity, special conditions can be applied.
Since the rods of the eyes are the visual receptors used in dim light, recording from the
rod system requires long periods of dark adaptation and one very dim single flash
stimulus. Using scotopic conditions and intense white flashes produces a mixed rod and
cone response. To obtain cone response, intense light flashes accompanied by an intense
light background are required (Good, 1999). The differentiation between rod and cone
activity is useful in the diagnoses of several diseases including inherited retinal
degenerations (mentioned above), which can cause night blindness. The degeneration
starts at more peripheral parts of the retina and progressively decreases the field of vision.
This disease can be recognized in the later stages by narrowing of the retinal vessels, and
by scattered clumps of pigment throughout the retina (Encyclopaedia Britannica, 1999).

7.10 Eye, ERG, EOG and Visual Field

19

Fig 7.14 The normal standardized corneal electroretinogram. S.F. is standard flash, and responses
to 1 log unit less than and greater than standard flash intensities are shown, as well as single flash
photopic and 30 Hz responses. Scale 100 V/div. and 20 ms per div. (Good, 1999)

7.10.2 Electro-oculogram (EOG)

The resting potential of the front of the retina is electropositive with respect to the back of
the retina. This makes the cornea positive compared to the back of the eye. If the eye
looks left, this positive cornea makes an electrode to the left of the eye more positive than
an electrode to the right of the eye. We can measure the differential voltage of about 100
V between the two electrodes to determine horizontal direction of gaze. Electrodes
above and below the eye can determine the vertical direction of gaze. This electrooculogram (EOG) is useful for measuring changes in direction of gaze, but cannot
measure absolute direction of gaze because variation in skin potential is larger than the
EOG signal. This test can be used to detect retinal pigment epithelial dysfunction
(Verdon, 2000).

20

Nervous System

7.10.3 Visual field

The peripheral vision of the retina reduces rapidly away from the fovea centralis (near the
center of the retina). This can be assessed by visual field measurement, in which the
examinee detects the targets that stimulate the retina at varying distances from the fovea.
Several methods are used to measure visual field, including the confrontation test and
methods using perimeters, tangent screens, or an Amsler grid (Newell, 1996).
Confrontation test
For this test the patient fixates their vision straight ahead on the examiner at a distance of
1 m. The patient closes one eye. The examiner puts one hand midway between the patient
and the examiner with some fingers extended. The examiner then slowly moves the hand
from the periphery toward the center and the patient states the number of fingers
displayed. The temporal and nasal fields and the superior and the inferior fields are tested
in turn.
Perimeters
A perimeter consists of a half bowl with a radius of curvature of 33 cm so that the eye is
at the center of rotation of the hemisphere.
In kinetic perimetry, the patient detects a test object of fixed size and
illumination moved from a nonsensing area. In static perimetry, the patient detects a test
object of constant size and location whose light intensity is increased until it can be seen.
An isopter, the contour line that connects the points at which an object of certain
size and certain light intensity may be recognized, represents a line on the macula lutea
that passes through the points of equal visual acuity.
Tangent screen
The evaluation of the central 30 of the visual field is desirable because the
photoreceptors of the eye are concentrated in and near the fovea centralis. This can be
conducted with a test object of size from 1 to 50 mm on a tangent screen, which contains
concentric circles every 5 and radiating lines 15 or 22.5 apart stitched or marked on
black cloth 1 m from the subject. This test demonstrates the peripheral isopter, the blind
spot, and various scotomas (blind spots in the visual field).
Amsler grid
An Amsler grid is formed by intersecting 21 horizontal and 21 vertical dark lines spaced
about 5 mm with one dark dot in the middle. It encompasses the central 20 of the visual
field when held at the reading distance. The patient tests an eye by fixating it on the

7.10 Eye, ERG, EOG and Visual Field

21

central dot and indicates any irregular line section that appears blurred, wavy, curved,
discolored, or missing. This may be an indication of leakage or bleeding in the back of
the eye causing retinal swelling (Freund, 1997). Amsler grid testing is recommended for
patients with macular degeneration or other retinal disordrs that threaten central vision. It
is advantageous because most patients are able to administer the test themselves, and if a
change in condition is noticed, they can come in to see their doctors immediately.
7.10.4 Eye pressure
Intraocular pressure is an important indication of glaucoma, a disease that can lead to
blindness. In a healthy eye, the aqueous humor is constantly produced by the ciliary body
in the eye and circulates into the chamber between the iris and the cornea from behind the
iris through the pupil. It can drain out of the eye through a network of tissues between the
iris and the cornea. If the network of tissues, called the drainage angle, does not function
well so that the aqueous humor flows away more slowly than it is produced, or even fails
to flow away, the pressure in the eye increases and causes extra pressure on the retinal
nerve fibers. Sometimes the extra pressure causes the collapse of tiny blood vessels that
nourish the light-sensitive cells of the retina and the optic nerve fiber. The cells and nerve
fibers begin to die and vision begins to fade in certain areas (Rich et al., 1996).
For the diagnosis of glaucoma, there are several important tests: tonometry is
used to measure the ocular pressure; gonioscopy is used to observe the anterior chamber
angle; ophthalmoscopy is used to evaluate the color and configuration of the cup and
neuroretinal rim of the optic disk (where the optical nerve enters the retina and is
therefore a blind spot); and perimetry is used to measure visual function in the central
field of vision.
Strictly speaking, the intraocular pressure cannot be measured unless we insert a
cannula into the eye. Clinically the intraocular pressure is measured indirectly by
measuring the ocular tension, which can be determined by the response of the eye to an
applied force. Two methods are used to measure ocular tension. Applanation tonometry
measures the force required to flatten a standard area of the cornea. Indentation
tonometry measures the deformation of the globe in response to a standard weight
applied to the cornea.
Figure 7.15 shows the Goldmann applanation tonometer, which measures the
force required to flatten the cornea by an area of 3.06 mm in diameter. The flattened area
is viewed with a biomicroscope through a split prism after instillation of a topical
anesthetic and fluorescein. We calculate the pressure p = f/a where f = force and a = area.
The Schitz tonometer measures the ease with which the cornea may be indented by the
plunger of the instrument. It is less costly and easier to use as compared with applanation
type instrument, but it is not as accurate as the applanation instrument and may be
influenced by more factors.
A noncontact applanation tonometer measures intraocular pressure without
touching the eye. An air pulse of linearly increasing force flattens the cornea in a few
milliseconds. When a beam of light reflects from the flattened surface, a maximum
detected signal shuts off the pneumatic pulse and measures the time elapsed, which
correlates with pressure.

22

Nervous System

The normal value of intraocular pressure is about 15 3 mmHg, while an eye

with glaucoma possibly has a intraocular pressure around 22 5 mmHg. The intraocular
pressure may vary with time by 6 mmHg during a day, so continuous sampling of the eye
pressure for several hours is required for adequate evaluation of glaucoma (Newell, 1996).

Examiners view
Biprism
Rod

Housing
Adjustment knob
Figure 7.15 The Goldmann applanation tonometer. When the applanation head is moved to the
right to rest on the anesthetized cornea, the examiner rotates the adjustment knob to increase the
force.

7.10.5 Ophthalmoscopy
Ophthalmoscopy is used to inspect the interior of the eye. It permits visualization of the
optic disk, vessels, retina, choroid (layer of blood vessels behind the retina), and ocular
media. Direct ophthalmoscopy and indirect ophthalmoscopy are the two methods used.
(Newell, 1996)
Direct ophthalmoscopy
In a darkened room, the examiner projects a beam of light from a hand-held
ophthalmoscope through the patient's pupil to view an upright image of the retina
structure. The ophthalmoscope has rotating lenses on top to magnify a particular area
being viewed up to 15 times. With pupillary dilation, about half the fundus (posterior part
of the eye) may be seen, while about 15% of the fundus may be seen without pupillary
dilation. The resolving power of direct ophthalmoscopy is about 70 m. Smaller objects,
such as capillaries, small hemorrhages, or microaneurysms, cannot be seen.

7.10 Eye, ERG, EOG and Visual Field

23

Indirect ophthalmoscopy
Indirect ophthalmoscopy, shown in Figure 7.16, is usually performed by using a
binocular ophthalmoscope. The patients retina is illuminated by a light source from
headset of the binocular instrument. An inverted image of the fundus magnified about 5
is formed between the condensing lens and the ophthalmoscope. Prisms within the
instrument make it possible to see a stereoscopic image. The entire fundus may be
examined by indirect ophthalmoscopy with pupillary dilation and sclera (the white part of
the eye) indentation.
Indirect ophthalmoscopy provides a significantly wider field of view than direct
ophthalmoscopy, but because of direct ophthalmoscopys higher resolution both are
commonly used during an eye examination (Bickford, 1995.) Although indirect
ophthalmoscopy only has 200 m resolving power, it is more advantageous than direct
ophthalmoscopy in that the stereoscopic image allows detection and evaluation of
minimal elevations of the sensory retina and retina pigment epithelium. These images
also allow the only direct view (without performing surgery) of the living network of
blood vessels and can help diagnose atherosclerosis, hypertension, diabetes mellitis, and
other systemic and eye-specific disorders (Bickford, 1995).

24

Nervous System

Mirror
Examiners eyes
Lamp

Inverted image of
patients fundus

Condensing lens

Patients eye ball

Figure 7.16 In indirect ophthalmoscopy, the examiner stands in front of the patient and holds the
condensing lens at an appropriate distance from the patients eye.

7.11 Ears and Audiometry

7.11.1 The ears
The ear is a sense organ specialized for hearing and balance functions. Anatomically, it is
divided into the external ear, middle ear, and inner ear. The external ear is composed of
the auricle, the external auditory canal, and the outermost layer of the tympanic
membrane. The middle ear, or tympanic cavity, is an oblong cavity bridged by an
ossicular chain consisting of three small bones between the tympanic membrane and the
inner ear. The inner ear consists of the cochlea and the labyrinth, which are the end organ
receptors for hearing and balance (Cody et al., 1981).

7.11 Ears and Audiometry

25

The auditory portion of the ear can be physiologically divided into two partsa
sound-conducting apparatus and an electromechanical transducer (Hawke, 1990). The
sound-conducting apparatus consists of the external ear, the tympanic membrane, the
ossicular chain, and the labyrinthine fluid. The electromechanical transducer transforms
the mechanical energy of sound into electric impulses to be transmitted by the auditory
nerves to the auditory cortex of the brain.
7.11.2 Audiometry
Audiometry is used to evaluate hearing pathologies in order to provide diagnostic
information and rehabilitation.
Pure tone air conduction threshold testing
The audio signal passes through the outer ear, the middle ear, and the inner ear before
being further processed by the central auditory system. Pathology of any of these parts of
the ear can cause hearing loss. In this test, the subject responds to bursts of singlefrequency stimuli presented through calibrated earphones. The signal intensity is adjusted
and the threshold with 50% correct responses is recorded (Laszlo and Chasin, 1988).
Pure tone bone conduction threshold testing
A special vibratory transducer is placed on the mastoid process or the forehead and thus
stimulates the inner ear directly through the skull. This bypasses the outer and middle
ears. If there is a difference between the air conduction and bone conduction responses,
this indicates pathology in the outer or middle ear, which may be treatable. Inner ear
pathology is difficult to treat (Laszlo and Chasin, 1988).
Speech discrimination testing
In this test, the subject listens to lists of single-syllable speech discrimination words
presented through earphones and repeats what he or she hears. The result of this test is
scored from 0 to 100% based on the correctness of the subjects answer. In contrast to the
pure tone threshold test, which addresses hearing sensitivity, this test assesses the
integrity of the entire auditory systems ability in hearing clearly and understanding
speech communication. A low score is related to sensorineural loss, and a higher score
may be attributed to normal hearing or conductive hearing loss (Laszlo and Chasin, 1988).
Speech reception threshold

26

Nervous System

This test uses words that are attenuated successively. The patient repeats back twosyllable words that have equal stress. A threshold is determined when the patient repeats
50% of the words correctly (Laszlo and Chasin, 1988).
Impedance audiometry
In tympanometry, a probe in the ear varies air pressure from +1962 to 1962 Pa testing
middle ear function and for the presence of fluid in the middle ear. A stiff middle ear
(poor ossicular chain mobility) reduces the mobility of the eardrum, causing a 220 Hz
tone to be reflected with little attenuation. A flaccid middle ear system (possible ossicular
chain disarticulation) yields high damping and little reflection (Laszlo and Chasin, 1988).
Evoked response audiometry (see section 7.2.3)
Clicks in the ear yield electric potentials measured by an electrode at the vertex (top) of
the scalp. An average response computer separates the very small synchronized signals
from the random signals from other brain potentials. Early latency evoked potentials
(about 10 ms) are from the brain stem. Middle latency responses (about 30 ms) are
responses above the brain stem. Late latency evoked potentials (about 100 ms) evaluate
the neurological integrity of the cortical centers.
7.11.3 Otoscopy
The external auditory canal and tympanic membrane, or eardrum, can be inspected with
an otoscope, a hand-held instrument with a flashlight, a magnifying glass, and a coneshaped attachment called an ear speculum. Evidence of injury or congenital (present at
birth) malformation is usually obvious. One can find foreign bodies, inflammation, ulcers,
lacerations, and tumors by visually inspecting the external auditory canal with a otoscope.
This type of ear examination can also detect a rupture or puncture of the eardrum. One
may also see scarring, retraction of the eardrum, or bulging of the eardrum. A red or
swollen eardrum can indicate an ear infection. In physical diagnosis, the otologist views
normal anatomy and pathologic processes in the middle ear through the translucent
tympanic membrane (Cody et al, 1981).

7.12 Muscles
7.12.1 Muscle contraction, length and force
About 40% of the human body is skeletal muscles, and another 10% is smooth muscles of
internal organs and cardiac muscles from the heart. Here we are interested in

7.12 Muscles

27

characterizing the function of skeletal muscles. The primary function of skeletal muscles
is to generate force. Because of this, they are excitable. Thus skeletal muscles have two
fundamental properties: They are excitable (able to respond to stimulus) and contractible
(able to produce tension) (Biewener, 1992).
A skeletal muscle consists of numerous fibers with diameters ranging from 10 to
80 m. Each muscle fiber contains hundreds to thousands of myofibrils (Figure 7.17).
Each myofibril has about 1500 myosin filaments and 3000 actin filaments lying side by
side (Carlson and Wilkie, 1974). Figure 7.18 shows the structure of a single myofibril.

Figure 7.17 The components of a skeletal muscle; the myofiber is the smallest complete contractile
system. Each myofiber is composed of many myofibrils. (from The Muscle Physiology Laboratory,
2000)

28

Nervous System

Myofibril

"Thin" actin
filament

"Thick" myosin
filament

Sarcomere

Z disk

Actin
lattice

Myosin
lattice

Anchoring
structure

Intermeshed
networks

Figure 7.18 The structure of a single myofibril. Each myofibril contains about 1500 myosin
filaments and 3000 actin filaments in the pattern shown above. The distance shown between the
two Z disks is one sarcomere. During muscle contraction the thin (actin) filaments slide inward past
the thick (myosin) filaments pulling the two Z disks closer together, thus shortening the sarcomere.
The shortening of the thousands of sarcomeres in each myofibril causes the myofibrils, the
myofibers and then the entire muscle to contract. (from The Muscle Physiology Laboratory, 2000)

As an action potential travels along a motor nerve to muscle fibers, it initiates an

action potential along the muscle fiber membrane, which depolarizes the muscle fiber
membrane and travels within the muscle fiber. The subsequent electrochemical reaction
within the muscle fiber then initiates attractive forces between the actin and myosin
filaments and causes them to slide together. This mechanism produces muscle contraction
(Jones and Round, 1990).
Tension is developed in the muscle as it contracts. There are three types of
contraction: Isometric, or static, contraction means a muscle contracts without change in
its length. Concentric contraction occurs when a load is less than the isometric force
produced by the muscle and the load shortens the muscle. Eccentric contraction occurs
when the load is greater than the isometric force and elongates the contracting muscle
(Sigel, 1986).

29

7.12 Muscles

The lengthforce relation of a muscle can be measured isometrically. With this

method, a skeletal muscle is maximally stimulated while its length is held constant. The
tension generated by the muscle at various constant lengths is measured to generate the
lengthforce curve (Lieber, 1992). Figure 7.19 shows a typical lengthforce curve of a
skeletal muscle (Ozkaya and Nordin, 1991).

Muscle

F
Fp

Fa

Muscle length
Resting length
Figure 7.19 The overall force of a muscle, F, is the sum of the active force, Fa, and the passive
force, Fp. The active force results from voluntary contraction of the contractile elements of the
muscle. The passive force results from elongation of the connective muscle tissue beyond its
resting length. No passive force builds up when the muscle is at its resting length or less.

The amount of force generated by a stimulated muscle depends on how its ends
are restrained (Woledge et al., 1985). If neither end of a stimulated muscle is fixed, the
muscle shortens at its maximum velocity of 33 cm/s, Vmax, and no force is generated. If
we fix one end of the muscle and apply a small force to the other end, the muscle
shortens at a steady velocity less than Vmax. A force of sufficient magnitude, F0, will
prevent the muscle from shortening and isometric contraction occurs within the muscle.
The forcevelocity relation is described by Hills equation (Woledge et al., 1985):
(F + a)(V + b) = (F0 + a)b

(7.2)

where a and b are constants derived experimentally, F is the muscle force during
shortening at velocity V, and F0 is the maximum isometric force that the muscle can
produce. Figure 7.20 shows a hypothetical curve for forcevelocity relation according to
Hills equation.

30

Nervous System

Shortening velocity

Vma

Muscle force, F/F0

F0
Figure 7.20 The muscle force decreases with increasing shortening velocity.

The forcevelocity curve can be obtained experimentally with isotonic

measurement (Lieber, 1992). The muscle is allowed to change its length against a
constant load during maximal stimulation. The muscle velocity (i.e., shortening velocity,
or lengthening velocity) is measured and plotted against the constant load.
7.12.2 Electromyography (EMG)
By placing electrodes into a skeletal muscle, we can monitor the electric activity of the
muscle (Keynes and Aidley, 1991). EMG is used to detect primary muscular disorders
along with muscular abnormalities caused by other system diseases such as nerve
dysfunction (Junge, 1992). Using EMG to study muscle function is also used in various
fields such as kinesiology, psychology, and rehabilitation medicine (Basmajian and De
Luca, 1985).
The two main types of electrodes for measuring EMG signals from muscles are
surface electrodes and inserted electrodes. The inserted type further includes needle
electrodes and wire electrodes. Selection of electrode type depends on the particular
application and the convenience of use (Loeb and Gans, 1986).
The surface electrode consists of silver disks that adhere to the skin. Saline gel
or paste is placed between the electrode and the skin to improve the electric contact. The
disadvantages of surface electrodes are that they cannot effectively detect signal from
muscles deep beneath the skin and that because of poor selectivity, they cannot eliminate
cross-talk from adjacent muscles.
Needle electrodes have a relatively smaller pickup area and thus are more
suitable for detecting individual motor unit potentials. Needle electrodes can be

7.12 Muscles

31

repositioned in the muscle after they are inserted. Figure 7.21 shows various
constructions of needle electrodes (Kimura, 1989).
(a)

(b)
(c)

(d)

(e)
Figure 7.21 Examples of needle electrodes: (a) Monopolar needle with exposed pickup tip. (b)
Monopolar configuration with a pickup wire centrally located in a hypodermic needle. (c) Bipolar
configuration with two pickup wires in parallel with each other in a hypodermic needle. (d)
Monopolar configuration with a wire exposed at needles side hole. This can be used to detect the
activity of individual muscle fibers. (e) Bipolar configuration with two wires exposed at the side
hole of a hypodermic needle. This can be used to detect the motor unit action potential from a large
portion of the motor unit territory.

Figure 7.22 shows a wire electrode. This electrode consists of two insulated fine
wires, 25 to 100 m in diameter, inserted through the cannula of a hypodermic needle.
The distal tips of the wires are deinsulated by about 1 to 2 mm and are bent to form
staggered hooks. One disadvantage of wire electrodes is that they tend to migrate after
insertion into the muscle.

Figure 7.22 A bipolar wire electrode uses a hypodermic needle to insert through the skin into the
muscle of interest.

We can obtain the EMG signal simply by placing a surface electrode on the skin
enveloping the muscle or by applying an inserted electrode in the muscle. A reference
electrode, usually a surface electrode, is placed on a site with minimal electric association
with the inserted site. The drawback of this monopolar configuration is that it detects not
only the signal from the muscle of interest but also unwanted signals from around the
muscle of interest.
In the bipolar configuration, two electrodes with a small distance between each
other are placed in the muscle to pick up the local signals within the muscle of interest. A
differential amplifier amplifies the signals picked up from the two electrodes with respect
to the signal picked up by a reference electrode. Because the interference signals from a

32

Nervous System

distant source are essentially equal in magnitude and phase as detected by the two
electrodes, the common mode rejection capability of the differential amplifier eliminates
the unwanted signals.
Figure 7.23 shows EMG waveforms of tongue muscle taken when the subject
was speaking some vowels.
Signal vs. number of points

x 10
x 10

Si gn al v s . Nu m ber o f Po in ts

Am pl i tud e

Amplitude

0
0

-5
-5

0
4
x 10

50
50

100

150
150

200
200

250

0
4
x 10

50

100

150

200
200

250

50

100
150
Number of points

200

250

x 10

10 0

25 0

Am pl itude

Amplitude

0
0

-5
-5

x 10

50

10 0

150

25 0

Am pl i tu de

Amplitude

0
0

-5
-5

50

10 0

150

200

25 0

Num ber of Poin ts

Figure 7.23 The EMG signal measured from the hyoglossus (tongue) muscle (b), styloglossus
muscle (c), and genioglossus muscle. Wire electrodes were inserted into these tongue muscles to
take the EMG measurement. The EMG signals are associated with the speech signal (a) of the
letters /ia/iao/ian/iang/iong.

7.12.3 Neuromuscular performance

Nerve cells that control movement emerge from the spinal cord and brainstem. They
travel various distances to reach particular muscles. Neuromuscular disorder develops
when any part along the path, from the brain to the muscle, does not work properly. If the
nerve that controls the muscle is injured and the action potential from the brain cannot
reach the muscle, the muscle does not contract normally. On the other hand, if the muscle
has some disease, it may not produce normal contraction in response to the action
potential from the brain.
For adequate treatment of neuromuscular disorders, it is important to know
where the injury occurs and how serious it is. By applying a dc stimulus to the skin over a
nerve at various positions in the legs or arms, and recording the time of muscle twitch
due to the stimulation with an electrode on the skin over the muscle, one can calculate the

7.12 Muscles

33

conduction velocity of the nerve. The conduction is slowed with injury to the nerve
(Ringel, 1987).
EMG is also used for diagnosis of neuromuscular performance. Abnormal
electric signals are recorded by the EMG if there is disease in the muscle.

7.13 References
American Academy of Family Physicans (AAFP). What You Should Know About
Epilepsy. [Online] http://familydoctor.org/handouts/214.html
Basmajian, J. V. and De Luca, C. J. 1985. Muscle Alive: Their Functions Revealed by
Electromyography. 5th ed. Baltimore, MD: Williams & Wilkins.
Bickford, L. 1995. The EyeCare Connection: The EyeCare Reports. [Online]
http://www.eyecarecontacts.com/exam.html
Biewener, A. A. (ed.) 1992. Biomechanics-Structures and Systems: A Practical Approach.
New York : IRL Press at Oxford University Press.
Campbell, N. A., Reece, J. B. and Mitchell L. G. 1999. Biology. 5th ed. Menlo Park:
Benjamin/Cummings.
Carlson, F. D. and Wilkie, D. R. 1974. Muscle Physiology. Englewood Cliffs, NJ:
Prentice-Hall.
Castello, R. B. (ed.) 1994. The American Heritage Dictionary. New York: Dell.
Clark, J. W. 1998. The origin of biopotentials. In Webster, J. G. (ed.) Medical
Instrumentation: Application and Design. 3rd ed. New York: John Wiley & Sons.
Cody, T. R., Kern, E. B. and Pearson, B. W. 1981. Diseases of the Ears, Nose, and
Throat. Mayo Foundation.
Early, P. J. and Sodee, D. B. 1995. Principles and Practice of Nuclear Medicine. 2nd ed.
St. Louis : Mosby.
Encyclopaedia Britannica. 2000. Eye disease: Retinal degeneration. [Online]
http://www.eb.com:180/bol/topic?eu=117507&sctn=18#s_top
Freund, K. B. 1997. Amsler Grid Testing of Central Vision. [Online]
http://www.vrmny.com/amsler.htm
Gevins, A. S. and Aminoff, M. J. 1988. Electroencephalography: Brain Electrical
Activity. In J. G. Webster (ed.) Encyclopedia of Medical Devices and
Instrumentation. New York: John Wiley & Sons.
Good, P. 1999. The Electroretinogram. [Online] http://www.eyenews.com/vol5_6.dir/features.dir/2feat5_6.html
Hari, R., and Lounasmaa, O. V. 2000. Neuromagnetism: tracking the dynamics of the
brain. Physics Today 3338, May 2000.
Hine, G. J. 1967. Instrumentation in Nuclear Medicine. New York: Academic.
Hughes, J. R. 1994. EEG in Clinical Practice. 2nd ed. Boston: Butterworth-Heinemann.
International Conference on Biomagnetism. 1989. Advances in Biomagnetism. New York:
Plenum Press.
Isley, M. R., Krauss, G. L., Levin, K. H., Litt, B., Shields, R. W. Jr. and Wilbourn, A. J.
1993. Electromyography/Electroencephalography. Redmond, WA: SpaceLabs
Medical.

34

Nervous System

Jones, D. A. and Round, J. M. 1990. Skeletal Muscle in Health and Disease : A Textbook
of Muscle Physiology. Manchester, New York: Manchester University Press.
Junge, D. 1992. Nerve and Muscle Excitation. 3rd ed. Sunderland, MA: Sinauer
Associates.
Keynes, R. D. and Aidley, D. J. 1991. Nerve and Muscle. 2nd ed. Cambridge, New York:
Cambridge University Press.
Kimura, J. 1989. Electrodiagnosis in Diseases of Nerve and Muscle: Principles and
Practice. 2nd ed. Philadelphia: Davis.
Laszlo, C. A. and Chasin, D. J. 1988. Audiometry. In J. G. Webster (ed.) Encyclopedia of
Medical Devices and Instrumentation. New York: John Wiley & Sons.
Lieber, R. L. 1992. Skeletal Muscle Structure and Function: Implications for
Rehabilitation and Sports Medicine. Baltimore: Williams & Wilkins.
Loeb, G. E. and Gans, C. 1986. Electromyography for Experimentalists. Chicago:
University of Chicago Press.
Malmivuo, J. and Plonsey, R. 1995. Bioelectromagnetism: Principles and Applications of
Bioelectric and Biomagnetic Fields. New York: Oxford University Press.
Mazziota, J. C. and Gilman, S. 1992. Clinical Brain Imaging: Principles and
Applications. Philadelphia: F. A. Davis Company.
Mettler, F. A. and Guiberteau, M. J. 1991. Essentials of Nuclear Medicine Imaging. 3rd
ed. Philadelphia: W. B. Saunders.
The Muscle Physiology Laboratory. 2000. Myofilament structure. [Online]
http://muscle.ucsd.edu/musintro/fibril.shtml
The Muscle Physiology Laboratory. 2000. Fiber types. [Online]
http://muscle.ucsd.edu/musintro/fiber.shtml
NATO Advanced Study Institute on Biomagnetism. 1983. Biomagnetism: An
Interdisciplinary Approach. New York: Plenum Press.
Newell, F. W. 1996. Ophthalmology: Principles and Concepts. 8th ed. St. Louis: Mosby Year Book.
Ozkaya, N. and Nordin, M. 1991. Fundamentals of Biomechanics: Equilibrium, Motion,
and Deformation. New York: Van Nostrand Reinhold.
Ringel, S. P. 1987. Neuromuscular Disorders: A Guide for Patient and Family. New
York: Raven Press.
Ritch, R., Shields, M. B. and Krupin, T. (eds.) 1996. The Glaucomas. 2nd ed. St. Louis:
Mosby.
Schneck, D. J. 1992. Mechanics of Muscle. 2nd ed. New York: New York University
Press.
Siedband, M. P. 1998. Medical imaging systems. In Webster, J. G. (ed.), Medical
Instrumentation: Application and Design. 3rd ed. New York: John Wiley & Sons.
Siegel, I. M. 1986. Muscle and its Diseases: An Outline Primer of Basic Science and
Clinical Method. Chicago: Year Book Medical Publishers.
Stein, R. B. (ed.) 1980. Nerve and Muscle: Membranes, Cells, and Systems. New York:
Plenum Press.
Van Heertum, R. L. and Tikofsky, R. S. (eds.) 1995. Cerebral SPECT Imaging. 2nd ed.
New York: Raven Press.
Vaughan, D. G. and Asbury, T. 1999. General Ophthalmology. 15th ed. Appleton &
Lange.

7.13 References

35

Verdon, W. 2000. Electrodiagnostic and Vision Functions Clinic. [Online]

http://spectacle.berkeley.edu/ucbso/vfc/
Wake Forest University Baptist Medical Center (WFUBMC). Diagnostic Neurology
Department. EEG Laboratory.
http://www.wfubmc.edu/neurology/department/diagneuro/eeglab.html
WFUBMC. Diagnostic Neurology Department. Evoked Potential Laboratory. [Online]
http://www.wfubmc.edu/neurology/department/diagneuro/ep.html
Webster, J. G. (ed.) 1998. Medical Instrumentation: Application and Design. 3rd ed. New
York: John Wiley & Sons.
Woledge, R. C., Curtin, N. A. and Homsher, E. 1985. Energetic Aspects of Muscle
Contraction. London; Orlando: Academic Press.
Yahoo! Health. 2000. Macular degeneration. [Online]
http://health.yahoo.com/health/Diseases_and_Conditions/Disease_Feed_Data/Macul
ar_degeneration/

7.14 Problems
7.1
7.2

7.3
7.4
7.5
7.6
7.7
7.8
7.9
7.10
7.11
7.12
7.13

7.14
7.15

Calculate the magnitude and sign of the cell membrane potential at body
temperature given that the intracellular potassium concentration = 150
millimoles/L and the extracellular potassium concentration = 5 millimoles/L.
Figure 7.1 shows the change of membrane potential due to an electric stimulus.
Suppose that the extracellular potassium concentration keeps constant during the
time course, sketch a figure to show the change of the intracellular potassium
concentration.
Calculate the velocity of propagation for the action potential in Figure 7.1.
Explain the origin of action potentials and how to measure them.
Explain why and how we measure the electroencephalogram.
Explain why and how we measure evoked potentials.
Draw the block diagram of a signal averager (average response computer) for
use with evoked potentials and explain how it improves the signal and by what
factor.
Calculate the improvement in signal-to-noise ratio when an average response
computer averages 200 recordings.
Explain the reason for an X-ray filter, collimator, grid, and screen.
Describe the beam paths used in a CT scanner.
Explain how magnetic fields are varied to obtain information from a single line
of tissue in an MRI scanner.
Describe the advantages of using nuclear medicine.
We wish to image the functionally active regions of the thyroid gland, which
takes up iodine when functionally active. Explain the procedure and equipment
used, including the detector type, localization method, and energy selection
method.
Explain how a gamma camera forms an image.
Explain why and how we diagnose visual system disease.

36
7.16
7.17
7.18
7.19
7.20
7.21

Nervous System
Describe the placement of electrodes to measure horizontal direction of gaze in
electro-oculography.
Explain why and how pressure within the eyeball is measured.
For the Goldmann applanation tonometer, calculate the force in newtons when
measuring a patient with normal ocular pressure.
Explain why and how we diagnose auditory system disease.
Explain why and how we measure dimension, force, and electrical activity of
skeletal muscle.
Design an experiment to determine the forcelength relationship of a muscle. Is
the method isometric or isotonic?

8
Heart and Circulation
Supan Tungjitkusolmun

Cardiovascular disease is the leading cause of death in the United States. There are over
one million heart attacks per year and over 600,000 deaths, of which 300,000 die before
reaching the hospital. One important trend that has emerged is the emphasis on prevention. Today, most people are aware of the things they can control that prevent heart disease (diet, not smoking, exercise, blood pressure, etc.). Attention to these areas helps reduce the chance of getting the disease but does not eliminate it completely. Early detection and treatment of the disease continue to be key areas of emphasis in the medical
community. The level of sophistication in dealing with cardiovascular diseases has taken
great strides in the past two decades and has been greatly aided by advances in technology. New diagnostic techniques have been developed and old ones improved, providing
added sensitivity and specificity. It is now very common for primary care physicians to
conduct tests in their offices to detect cardiovascular diseases.
It is very important for biomedical engineers to understand the basic principles
of medical instruments. This chapter presents several important cardiovascular variables
and the most common methods by which they are measured, as well as an introduction to
the physiology of the heart and blood vessels.
The human heart and blood vessels are a transportation system that delivers essential materials to all cells of the body and carries away the waste products of metabolism. The heart serves as a four-chamber mechanical pump. The two chambers on the
right side of the heart send deoxygenated blood to the lungs via the pulmonary arteries
(pulmonary circulation). The other two chambers, on the left side of the heart, supply
oxygenated blood through the arteries, which branch in an orderly fashion until they
reach the capillary beds of the organs (systemic circulation). Here, there is an exchange
of nutrients, waste products and dissolved gases. The blood then flows from the capillaries into the veins, which lead back to the right atrium. Hormones are also transported by
the cardiovascular system.

8.1 Cardiac Anatomy and Physiology

Heart and Circulation

Figure 8.1 illustrates important features of the human heart. The heart, about the size of a
clenched fist, is located in the thoracic (chest) cavity between the sternum (breastbone)
and the vertebrae (backbone). The heart is divided into right and left halves and has four
chambersan upper and lower chamber within each half. The upper chambers, the atria,
receive blood returning to the heart and transfer it to the lower chambers, the ventricles,
which pump the blood from the heart. Figure 8.2 shows a simplified circulatory system.
Blood returning from the systemic circulation enters the right atrium via large veins
known as the venae cavae. The blood entering the right atrium has returned from the
body tissues, where O2 has been extracted from it and CO2 has been added to it by the
cells of the body. This partially deoxygenated blood flows from the right atrium into the
right ventricle, which pumps it out through the pulmonary artery to the lungs. Thus, the
right side of the heart pumps blood into the pulmonary circulation. Within the lungs, the
blood loses its CO2 and picks up a fresh supply of O2 before being returned to the left
atrium via the pulmonary veins. This highly oxygenated blood returning to the left atrium
subsequently flows into the left ventricle, the pumping chamber that propels the blood to
all body systems except the lungs; that is the left side of the heart pumps blood into the
systemic circulation. The large artery carrying blood away from the left ventricle is the
aorta.

Figure 8.1 Basic structure of the heart. RA is the right atrium, RV is the right ventricle; LA is the
left atrium, and LV is the left ventricle. Basic pacing rates are shown.

8.2 Cardiac Biopotentials

8.1.1 Heart valves

Blood flows through the heart in one fixed directionfrom veins to atria to ventricles to
arteries (see Figure 8.2). The presence of four one-way heart valves aides the unidirectional flow of blood. The valves are positioned so that they open and close passively due
to pressure differences, similar to a one-way door. The backward gradient can force the
valve to close but cannot force it to swing open in the opposite direction.

Deoxygenated
blood

Oxygenated
blood

Upper body

Right
atrium

Lung

Right
ventricle

Left
atrium

Left
ventricle
Lower body

Figure 8.2 The simplified circulatory system. The blood is delivered from the right ventricle to the
lung. The oxygenated blood from the lung is then returned to the left atrium before being sent
throughout the body from the left ventricle. Deoxygenated blood from the body flows back to the
right atrium and the cycle repeats.

Two of the heart valves, the right and left atrioventricular (AV) valves, are positioned between the atrium and the ventricle on the right and left sides, respectively (see
Figure 8.1). These valves allow blood to flow from the atria into the ventricles during
ventricular filling (when atrial pressure exceeds ventricular pressure), but prevent the
backflow of blood from the ventricles into the atria during ventricular emptying (when
ventricular pressure greatly exceeds atrial pressure). If the rising ventricular pressure did
not force the AV valves to close as the ventricles contracted to empty, much of the blood
would inefficiently be forced back into the atria and veins instead of being pumped into
the arteries. The right AV valve is also called the tricuspid valve because it consists of
three cusps, or leaflets. Likewise, the left AV valve, which consists of two leaflets, is
often called the bicuspid valve or, alternatively, the mitral valve.

Heart and Circulation

The other two heart valves, the aortic and pulmonary valves, are located at the
junction where the major arteries leave the ventricles. They are also known as semilunar
valves because they are composed of three leaflets, each resembling a shallow half-moon
shaped pocket. These valves are forced open when the left and right ventricular pressures
exceed the pressure in the aorta and pulmonary arteries (i.e. during ventricular contraction and emptying). The valves close when the ventricles relax and ventricular pressures
fall below the aortic and pulmonary artery pressures. The closed valves prevent blood
from flowing from the arteries back into the ventricles from which it has just been
pumped. The semilunar valves are prevented from swinging back into the ventricles by
the anatomical structure and positioning of the cusps (Sherwood, 2001).
8.1.2 Cardiac cycle
A cardiac cycle contains a sequence of events during one heart beat. The ventricles contract during systole. The ventricles fill during diastole. In an average human at rest, the
entire cardiac cycle takes about 0.8 s (giving a pulse of about 75 beats/min). Heart rate is
the number of heartbeats per minute. We can easily measure our own heart rate by counting the pulsations of arteries in the wrist (radial artery) or neck (carotid artery). Figure 8.3
shows the synchronization of the electrocardiogram (see section 8.2.2) and the pressures
in the left ventricle.

8.2 Cardiac Biopotentials

Figure 8.3 In the top figure, the electrocardiogram (ECG) initiates the cardiac cycle. The cardiac
sounds (section 8.5) are also shown. The bottom figure shows that ejection occurs when the pressure in the left ventricle exceeds that in the arteries.

During diastole, the pressure in the left ventricle is low and less than that of the
arteries. Thus the aortic valve is closed. The blood from the left atrium flows into the left
ventricle. The left atrium subsequently contracts due to the excitation from the pacemaker
cells in the heart (see section 8.1.3) and further fills the left ventricle. The ventricles contract, which increases the pressure and closes the mitral valve. The pressure increases
until it exceeds the aortic pressure, the aortic valve opens, and blood flows into the aorta.
The blood continues to flow from the left ventricle into the aorta as long as the ventricular pressure is greater than the arterial pressure, as shown in Figure 8.3. However, if the
venticular pressure is much greater than the aortic pressure, there is a problem with the
valve and it may be stenotic (narrow). The dicrotic notch results when the aortic valve
slams shut and the AV valve is still closed. Table 8.1 summarizes the events of the cardiac cycle.

Heart and Circulation

Table 8.1 Duration and characteristics of each major event in the cardiac cycle.
Event

Characteristics

Duration at 75
bpm (0.8 s cycle)

Atrial diastole
Ventricular diastole

AV valves opened.
Semilunar valves close.
Ventricular filling.
AV valves open.
Semilunar valves closed. Ventricular filling.
AV valves closed.
Semilunar valves open.
Blood pumped into aorta and pulmonary artery.

0.4 s

Atrial systole
Ventricular diastole
Atrial diastole
Ventricular systole

0.1 s
0.3 s

The aortic pulse contour changes with aging due to changes in aortic compliance
(distensibility). Reflection of the systolic pulse from the distal aorta distorts the aortic
contour and creates two systolic shoulders. A pulse from a younger person with a more
compliant aorta is rounded. That from an older person with decreased aortic compliance
has a reflected wave, which causes a more peaked pulse and higher pressure (hypertension) (ORourke, 1995).
Currently, there are numerous electric heart rate monitoring systems that continuously measure heart rate during rest and exercise. They rely on sensing electrodes
placed on the skin surface, which measure the electric frequency of the heart, or the electrocardiogram (section 8.2.2). Some models, like the ear clip, use a photosensor to measure change in tissue light absorption caused by the pulsatile increase in arteriole volume,
which is then converted into a beat-per-minute readout.
8.1.3 Cardiac excitation and control
Figure 8.1 shows that the sinoatrial node (SA node), which serves as the hearts pacemaker, is located in the wall of the right atrium, near the point where the inferior vena
cava enters the heart. It is composed of specialized muscle tissue that combines characteristics of both muscle and nerve (see Chapter 7). Nodal tissue contracts like muscle, thus it
generates electric impulses similar to those found in nervous tissue. This tissue is selfexcitable, and can contract without any signal from the nervous system. Each time the SA
node contracts, it initiates a wave of excitation that travels through the wall of the heart at
a rate of approximately 1 m/s (Berne, 1981). The impulse spreads rapidly, and the two
atria contract simultaneously. Cardiac muscle cells are electrically coupled by the intercalated disks between adjacent cells. At the bottom of the wall separating the two atria is
another patch of nodal tissue, the atrioventricular (AV) node. The atria and ventricles are
separated by nonconductive tissue except for the AV node. When the wave of excitation
reaches the AV node, it is delayed for about 0.1 s (AV-nodal delay), which ensures that
the atria will contract first before the ventricles contract. After this delay, the signal to
contract is conducted to the apex of the ventricles along the bundle of His, and the wave
of excitation then spreads upward through the ventricular walls via the Purkinje fibers
(Davies et al., 1983).

8.2 Cardiac Biopotentials

Simple tests of the functions of the heart include monitoring the heart rate, the
blood pressure, the cardiac rhythm, and the heart sounds. More sophisticated diagnoses
can be performed by studies of the electrocardiogram, the echocardiogram, the stress test,
Holter monitoring, and angiography. These techniques will be discussed in the subsequent sections.

8.2 Cardiac Biopotentials

8.2.1 Electrodes
Surface recording electrodes convert the ionic current within the body to electron current
in the metal wires, which is then sent to a recorder. Electrodes are generally made from
metal. The silver/silver chloride (Ag/AgCl) electrode is electrochemically stable and is
thus most widely used (Brown et al., 1999). An electrode is usually constructed by electrolyzing a silver plate as an anode in an aqueous solution of sodium chloride to form a
film of AgCl on the surface of the silver. The reaction is Ag + Cl AgCl + e. Often a
transparent electrolyte gel containing chloride ions as the principle anion is used. The gel
also allows for good contact between the skinelectrode interface. Figure 8.4 shows a
disposable surface electrode. Tobuchi et al. (1981) discuss several techniques used to
manufacture Ag/AgCl electrodes. Other metals such as nickel-plated brass can be used as
reusable ECG electrodes but they have larger motion artifacts.

Figure 8.4 A disposable surface electrode. A typical surface electrode used for ECG recording is
made of Ag/AgCl. The electrodes are attached to the patients skin and can be easily removed.

Motion artifacts (error caused by movement of the electrodes) contribute a large

error to signals from surface electrodes. Artifact may be caused by stretching of the skin,
which changes the voltage across the skin. This can be minimized by abrading the skin
with sandpaper, which short circuits the voltage. There are several types of electrodes:
suction electrodes, floating metal body-surface electrodes, and dry electrodes (Neuman,
1998a). Other types of electrodes, such as microelectrodes, are discussed in Chapter 7.

Heart and Circulation

8.2.2 Electrocardiogram
Contraction of cardiac muscle during the heart cycle produces electric currents within the
thorax. Voltage drop across the resistive tissue can be detected by electrodes placed on
the skin and recorded as an electrocardiogram (ECG, sometimes called EKG). Figure 8.3
shows components of an electrocardiogram (P, Q, R, S, and T waves).
The electrocardiogram contains information on the electric rhythm, electric conduction, muscle mass, presence of arrhythmia (irregular heart beat), ischemia (lack of
blood flow) or infarction, even electric disturbance and drug effects on the heart (see Figure 8.5). The P wave in the ECG represents the contraction of the atria. From Figure 8.3,
the R wavealso called the QRS complexof the ECG corresponds to the contraction
of the ventricles. The T wave is due to repolarization (re-establishing the electric potential) of the ventricles. The atria repolarize also, but this event is covered up by the QRS
complex. The downslope of the T wave is referred to as the vulnerable period. It is during
this time of the cardiac cycle that a premature impulse may arrive and fibrillation can
occur. Fibrillation is an irregular contraction of cardiac muscle, which results in an ineffective propulsion of blood. When ventricular fibrillation occurs a normal rhythm can be
re-established by passing a strong electric current. This puts the entire myocardium in a
refractory state in which no impulse can occur. Thus, resetting the heart. Direct current
defibrillation shock has been found to be more effective than alternating current shock
(Berne, 1981).

aVR

V1

V4

(a)

aVR

V1

V4

(b)
I

V1

aVR

(c)

V4

8.2 Cardiac Biopotentials

I

aVR

V1

V4

(d)

aVR

V1

V4

(e)
I

aVR

V1

V4

(f)
Figure 8.5 The electrocardiogram. (a) The normal ECG. (b) 1st degree AV block in which the
delay from the P wave to the Q wave is lengthened. (c) acute inferior myocardial infarction (lack of
blood flow to heart muscle, which causes tissue to die), in which the ST sement is depressed. (d)
right atrial hypertrophy (increase in muscle mass of the atria), in which V4 is large. (e) ventricular
tachycardia (faster than normal heart rate) with clear AV dissociation. (f) WolffParkinsonWhite
syndrome with atrial fibrillation.

The ECG recording instrument is called an electrocardiograph. Figure 8.6

shows the block diagram of a typical electrocardiograph. The leads attached to the surface electrodes from the patient are fed to the resistors and switch network in order to
select the desired potential difference between electrodes. The input signals are amplified
and converted from analog signals to digital signals before being processed by a microprocessor. Several algorithms within the microprocessor perform QRS detection and filtering. Clinicians can view the results on the screen of a microcomputer or print a hard
copy of the ECG signals for further analysis. Onoda (1994) shows the complex system of
a modern electrocardiograph. An electrocardiograph provides the physician with valuable
information on various physical characteristics of the heart, the extent and progress of
damage incurred by the heart, and the influence of drugs on the patient.

10

Heart and Circulation

RA

Figure 8.6 Block diagram of an electrocardiograph. The normal locations for surface electrodes are
right arm (RA), right leg (RL), left arm (LA), and left leg (LL). Physicians usually attach several
electrodes on the chest of the patients as well.

Clinical electrocardiography would not be practical without the differential amplifier. The differential amplifier rejects ground-referred interference. Figure 8.7 shows
an ECG amplifier with a dc-coupled instrumentation amplifier, which has a gain of 25.
The high-pass filter blocks dc offset and the last stage is a noninverting amplifier which
has a gain of 34 (Neuman, 1998b). Chapter 2 provides more information on the differential amplifier.

Figure 8.7 A circuit of an ECG amplifier. The instrumentation amplifier, located on the left of the
circuit provides a high input impedance and has a gain of 25 in the dc-coupled stages. (From Webster, J. G. (ed.) 1998. Medical instrumentation: application and design, 3rd ed. New York: John
Wiley & Sons.)

8.2 Cardiac Biopotentials

11

Connecting each lead of the ECG to a buffer amplifier prevents loading error
caused by the added resistance of the skin. The buffer amplifier offers a very high input
impedance and unity gain. If a differential amplifier or inverting amplifier were used
without a buffer, the gain would be affected (see section 2.2.5).
12-lead ECG
The electric potential generated by the heart appears throughout the surface of the body.
We can measure the potential differences between surface electrodes on the body. Different pairs of electrodes at different locations generally yield different results because of
the spatial dependence of the electric field of the heart (Neuman, 1998b).
Physicians attach several surface electrodes and record a few beats of the 12lead diagnostic ECG (see Figure 8.6). The 12-lead ECG monitors the hearts electrical
activity from 12 different angles. By selecting different pairs of electrodes, we can obtain
ECGs that have different shapes. Common locations for the electrodes are the
limbsright arm, left arm, right leg, and left leg. A pair of electrodes, or combination of
several electrodes through a resistive network that gives an equivalent pair, is referred to
as a lead. The most widely used bedside ECG diagnostic system records 12 different potential differences, or ECG leads (Tompkins, 1993).
Figure 8.8 shows the Einthoven triangle. The three leads shown are lead I, right
arm (RA) to left arm (LA); lead II, right arm (RA) to left leg (LL); and lead III, LA to LL.
The scalar signal on each lead of Einthovens triangle can be represented as a voltage
source, thus we can write Kirchhoffs voltage law for three leads as

I II + III = 0

(8.1)

12

Heart and Circulation

LA

RA

LL
Figure 8.8 Einthovens triangle. Lead I is from RA to LA, lead II is from RA to LL, and lead III is
from LA to LL.

Three additional leads, called the unipolar leads, or augmented limb leads, are
routinely used in taking diagnostic ECGs. These leads are based on signals obtained from
more than one pair of electrodes. Unipolar leads consist of the potential appearing on one
electrode taken with respect to an equivalent reference electrode, which is the average of
the signals from two or more electrodes. The three different unipolar leads are positive
electrodes placed on the right arm, augmented right (aVR), the left arm, augmented left
(aVL), and the left foot, augmented foot (aVF). The other leads that are frequently measured in clinical ECGs are the precordial leads. Physicians place an electrode at various
anatomically defined positions on the chest wall (Neuman, 1998b). Required frequency
response is 0.05 to 150 Hz.
ECG monitoring

The ECG may be monitored continuously when the patient is in emergency care, in a
coronary care unit, an intensive care unit, or during stress tests, by a recorder or telemetry.
In these cases, only one lead, usually lead II, is monitored on a display. The system
automatically monitors rhythm disturbances and gives an alarm when the heart rate is too
slow or too fast. To avoid motion artifacts and muscle noise the frequency response is
reduced to 0.5 to 40 Hz.
Ambulatory (Holter) monitor

8.2 Cardiac Biopotentials

13

An ambulatory monitor, also called a Holter monitor, is a portable electrocardiograph,

which can be worn for up to 24 h to monitor heart abnormalities. The monitor records the
hearts activity through electrodes placed on the chest of a patient. The electric impulses
are then transmitted to an amplifier, which records them on a small recorder for later review by a physician. The recording provides a doctor with important information about
the heart and its rhythm and can help identify the cause of such normally encountered
symptoms as chest pain, palpitations (abnormally rapid pulsation in the heart), or dizziness.
Ambulatory electrocardiography has been available since the early 1950s, when
Dr. Holter introduced his portable electrocardiograph to the medical community. Since
that time, these devices have decreased in size and weight, but have increased in sophistication. The first devices had to be worn in a backpack, and despite their large size and
weight, could only review the heart along one axis. Modern devices can look at the heart
along several different axes, and their weight and size are such that they do not interfere
with the patients daily activities.
The electrodes are applied to the chest. The number of wires and electrodes applied depends upon the number of leads the physician feels should be monitoredusually
one or two. After the electrodes are securely attached to the chest, the excess wire is
taped to the skin to prevent accidental disconnection. After the technician checks the system to make sure that it is working properly, the monitor is connected to an electrocardiogram recorder to determine if the ambulatory electrocardiogram provides a high quality tracing, free of excessive artifact or electric interference. The recorder is then tested
and a battery inserted into the device. Patients can wear the recorder either on their belt or
over their shoulder.
Two types of Holter monitors are generally used. The first type is the tape-based
system. This system is composed of three components: a monitor with a cassette tape,
which records the ECG signals, a PC scanning device that scans at 60 times real time and
uses a data-reduction processor, and a printer. The patient wears the monitor, which records an analog recording on the cassette tape. When the monitoring session is finished,
the cassette tape is put into the data-reduction processor and data are digitally stored on
the hard drive of the PC. From the PC, a report is generated, reviewed, and printed. These
reports are in two or three channel format depending upon the instrumentation. Tapebased systems are usually preferred by physicians who need numerous recorders because
these are less expensive than digital recorders.
The second type of monitor is the digital solid state system. This system is usually made up of two components, the monitor and the printer. In some cases, a PC is used
to visualize the information on a screen. The patient wears the solid-state monitor, which
digitizes, stores, and analyzes the ECG signal in real time. In a tape-based system, the
computer performs the function of digitizing, storage and analysis. Digital Holters have
no moving parts or tapes. They produce a two- or three-channel report depending upon
the instrument and are generally more expensive than individual tape-based monitors
because of their capabilities.
ECG telemetry system

14

Heart and Circulation

Medical personnel usually monitor one lead of the patients ECG using a bedside monitor
in the critical care unit (CCU) of a hospital. As the patient recovers, the doctor may feel
comfortable with the patient moving around but may want to maintain ECG monitoring.
A bedside monitor limits the movement of the patient to the length of the ECG leads.
With a telemetry system, a nurse attaches the patient leads to the patient in the same way
as for the bedside monitor. Instead of the patient leads feeding back directly to the bedside monitor, they feed a portable transmitter. The transmitter, which resembles a portable radio, is small enough for the patient to carry. The transmitter sends the patients
ECG by modulating it with a radio-frequency (RF) carrier, much in the same way that a
cordless telephone works. Manufacturers may transmit the information using frequency
modulation (FM), amplitude modulation (AM) or even pulse code modulation (PCM)
depending upon design. Antenna systems pick up the RF signal from the transmitter and
send it to the receiver. The receiver then removes the RF signal from the patients ECG
and sends the resulting signal to the monitor for analysis. The signal strength of the
transmitter and tuning of the system limit the distance a patient can move away from the
antenna system. In the ECG telemetry system, the receiver alarms on high and low ECG
limits the user can adjust, or possibly some arrhythmia conditions (Reinhard and Fowler,
1987).
Exercise stress testing

Exercise stress testing is an evaluation of the patients cardiovascular system using one
lead of an ECG, monitor, treadmill, and blood pressure unit. It is done to screen patients
for heart disease and help predict or unmask potential coronary problems.
The procedures for stress testing are as follows. The technician first abrades the
skin and applies stress electrodes to the patients chest. Then, a baseline ECG and blood
pressure are taken. The exercise protocol (there are numerous protocols, which gradually
increase both speed and elevation of the treadmill) is started. During exercise, the physician monitors the patients ECG, heart rate, blood pressure, and physical symptoms. ST
segment depression may indicate cardiac ischemia due to coronary arterial occlusion.
When the test ends, ECG and blood pressure measurements are done for documentation
of the heart rate at maximum effort or submaximum effort. It is important to continue
monitoring the patients ECG, heart rate, and blood pressure during recovery, because
some problems are uncovered during that phase.
8.2.3 Electrograms

Electrograms are electric signals recorded from within the heart. Electrograms are measured by introducing a catheter electrode into the veins. The catheter is then placed into
the heart under X-ray guidance, which is usually a biplane cardiac imaging system with a
C-arm, allowing multidirectional biplane projections with extreme angulation capabilities.
We can use electrograms to study abnormal heart rhythms under controlled situations to
diagnose the specific problem with the hearts electric system. Electrophysiologists can

8.2 Cardiac Biopotentials

15

obtain information on the propagation of excitation from the SA node to the ventricles
with the catheter.
There are certain groups of patients who have problems that can be corrected by
radio-frequency ablationthe use of RF energy via a catheter for lesion making to modify the conduction pathway of the heart (Huang and Wilber, 2000). One example is patients with refractory arrhythmias of the upper chamber who need complete interruption
of the AV conduction pathway. Electrophysiologists usually monitor the abnormal conduction of the heart by applying three ablation catheters, each having three recording
electrodes and one ablation electrode at the tip. The catheters are placed at different locations, such as the SA node, the AV node, and in the ventricle. By observing the timing of
the excitation activity at various locations, the electrophysiologist can determine the locations of unwanted conduction pathways. They are then heated and destroyed with RF
energy.
Any invasive procedure that requires catheters to be inserted in the body includes some risk. However, the risk is quite small and the electrophysiology study is relatively safe. One possible side effect is that some patients may develop bleeding at the site
of catheter insertion into the vessels.

8.3 Cardiac Pressures

The state of the heart and heart valves can be assessed by measurement of cardiac pressures. The sensors used are designed so that the application of pressure modifies the electric properties of some component by stretching or otherwise deforming it. Sensors with
this property are called strain gages. Strain gages are very important in biomedical instrumentation design. Section 8.3.1 describes the fundamentals of strain gages. Section
8.3.2 describes a catheter that is widely used for invasive measurements of cardiovascular
variables.
Figure 8.9 shows a typical system for measuring cardiac pressures. A pressure
port is located at the tip of the catheter. A syringe temporarily inflates a balloon to guide
the catheter into the pulmonary artery. Slow infusion prevents a clot from forming at the
pressure port.

16

Heart and Circulation

Pressure port
Catheter

Figure 8.9. A system for cardiac pressure and flow measurement. The pressure is usually measured
with a catheter placed in the right side of the heart. An external strain gage pressure sensor is also
shown. (Adapted from Orth, J. L. 1995. System for calculating compliance and cardiac hemodynamic parameters, US Patent, 5,423,323.)

8.3.1 Strain gage

Pressure measurement systems in use today are essentially all electric strain gages and
apply the principle of the Wheatstone bridge, as described in sections 2.1.6 and 9.11.3.
Older pressure sensors had unbonded strain gage wires mounted under stress between a
frame and movable armature so that preload was greater than any expected external compressive load. This was necessary to avoid putting the wires in compression. Thus blood
pressure caused diaphragm movement, which caused resistance change, which caused a
voltage change.
Modern pressure sensors use an etched silicon diaphragm embedded with strain
gages formed by diffusing impurities into it. The strain gage electric resistance changes
with diaphragm deformation, which responds to pressure applied. Unfortunately the resistance of the silicon also changes with temperature, but the effect of this change may be
minimized by diffusing four strain gages into the diaphragm. These then form the four
arms of a resistance bridge. Two of the strain gages are placed close to the center of the
diaphragm, where their resistance increases with applied pressure, and the other two close
to the periphery, where their resistance decreases with pressure. However, if the temperature changes, all four resistances change by the same percentage and this does not change
the output from the resistance bridge.

8.3 Cardiac Pressures

17

8.3.2 Catheter

In the cardiac catheterization laboratory, catheters are inserted into the chambers of the
heart to measure pressures, flows, and oxygen saturation to determine if valve replacement is required. Catheters can also inject radiopaque dye for X-ray fluoroscopy, which
can image vessel narrowing to determine if vessel replacement is required.
A catheter is a flexible tube for insertion into a narrow opening such as the
blood vessels, so that fluids may be introduced or removed. Figure 8.9 shows a Swan
Ganz catheter. The first person who passed a catheter into the human heart was Werner
Forsmann, a medical student at Eberswalde, Germany. In 1929, at the age of 25, he
passed a 65 cm catheter through one of his left antecubital veins, guiding it by fluoroscopy until it entered his right atrium. Forsmann then climbed up the stairs to the Radiology Department where the catheter position was then documented.
Catheters are inserted through sheaths (the layers of connective tissue that envelop structures such as nerves, arteries, tendon, and muscle) into the arteries and the
veins of the body then pass up to the heart. The pressures within different areas of the
heart are measured. Two approaches may be used to apply a catheter to the heart.
The brachial (in the arm) approach usually utilizes cutdown on the brachial
artery (an artery that extends from the auxillary artery, down the side and inner surface of
the upper arm to the elbow, where it divides into the radial and ulnar arteries) and vasilic
vein at the elbow. The direct brachial approach may have advantages in a very obese patient, in whom the percutaneous femoral technique may be technically difficult and
bleeding hard to control after catheter removal.
The indirect femoral approach. The percutaneous femoral approach has its own
set of advantages. Arteriotomy and arterial repair are not required; it can be performed
repeatedly in the same patient at intervals, whereas the brachial approach can rarely be
repeated more than two or three times with safety.
Cardiac pressures are usually measured by one of several ways. In the cathetertype system, the blood pressure communicates with the pressure sensor via a fluid-filled
catheter. The sensor is outside the body. With an appropriate length and diameter, accurate pressure readings can be obtained.
With the catheter-tip sensor, the elastic unit is placed in a blood vessel, thus the
elastic member is in direct contact with the blood. This arrangement avoids the damping
and resonance sometimes encountered with catheter-type systems and is more accurate
than the catheter-type sensors (Grossman and Baim, 1991).

8.4 Cardiac Output

Cardiac output is a measure of the well being and performance of the heart. The maintenance of blood flow commensurate with the metabolic needs of the body is a fundamental
requirement of human life. In the absence of major disease of the vascular tree, the maintenance of appropriate blood flow to the body depends largely upon the hearts ability to
pump blood in the forward direction. The quantity of blood delivered to the systemic cir-

18

Heart and Circulation

culation per unit time is termed the cardiac output, generally expressed in liters/minute
(L/min).
Cardiac output of a normal patient is directly correlated with the size of the body.
Most investigators use total body surface area as the standardizing variable. The ratio of
the cardiac output to area is called the cardiac index. Table 8.2 lists several variables related to cardiac output.
Table 8.2 Some physiological variables. The data presented in this table are the average values of a
group of subjects.
Variables

Mean (
SD)

Weight (kg)
Cardiac output (mL/s)
Heart rate (min1)
Mean velocity, ascending aorta (mm/s)
LV end-diastolic volume (mL)
LV end-systolic volume (mL)
LV stroke volume (mL)
LV ejection fraction
LV mean wall thickness (mm)

70
110
76
16
125 (31)
42 (17)
82 (20)
0.67 (0.11)
10.9 (2.0)

Cardiac output (CO) = heart rate (HR) stroke volume (SV)

(8.2)

The ejection fraction is defined as

Ejection fraction =

Stroke volume
End - diastolic volume

(8.3)

End-diastolic volume is the amount of blood in the ventricle at the end of diastole. End-systolic volume is the volume of blood remaining in the ventricle at the end of
systole when ejection is complete.
The stroke volume, the volume of ejected blood from the ventricles, is about 80
mL/beat. The average resting heart rate of human is about 70 beats/min. Thus cardiac
output is about 80 70 = 5,600 mL/min = 5.6 L/min. Cardiac output is regulated by
changes in both HR and SV. Heavy exercise increases both HR and SV, and CO can increase to as high as 25 L/min.
8.4.1 Fick method

In 1870, Fick stated that if the concentration of O2 in arterial and mixed venous blood is
known, and the rate of inhalation of O2 is also known, the cardiac output, CO, can be
computed from the formula (Webster, 1998)
CO =

dm dt
Ca Cv

(8.4)

8.4 Cardiac Output

19

where dm/dt is the consumption of O2 (L/min), and Ca is the arterial concentration of O2

(L/L). Cv is the venous concentration of O2 (L/L).
The Fick technique is invasive, that is, the O2 concentrations must be measured
in the vicinity of the lungs. The rate of inhalation or exhalation of gas is measured using
the spirometer as discussed in Chapter 9. Ca can be measured by obtaining a sample from
any artery (except the pulmonary artery) and Cv can be measured in the pulmonary artery.
The blood-O2 analyzer is used to find the concentrations of O2 in blood.
The Fick method has been widely used. Its accuracy when carefully performed
is approximately 5%. The principal limitation is the need for a steady state; output and
pulmonary gas exchanged must remain constant while samples are collected, which in
practice means a period of several minutes. For that reason, rapid changes in the circulation during exercise or other events cannot be studied by this technique.
Example 8.1 A patients O2 concentration, meaused in the pulmonary artery, is 0.12 L/L.
The O2 concentration measured in the patients aorta is 0.19 L/L. A spirometer is used to
obtain the patients O2 consumption rate 0.250 L/min. Calculate the patients cardiac
output.
The patients O2 concentrations and consumtion are known: Ca = 0.19 L/L, Cv
= 0.12 L/L and dm/dt = 0.250 L/min. From Eq. (8.4)

CO =

dm dt
0.250 L/min
=
= 3.57 L/min
Ca C v 0.19 L/L 0.12 L/L

8.4.2 Thermodilution

Thermodilution is considered a gold standard for measuring cardiac output. A bolus of 10

mL, 0.9% cool physiological saline is injected into the venous circulation in the right
atrium. The drop in temperature in the arterial circulation is usually measured using a
thermistor attached to a catheter in the pulmonary artery. The cardiac monitor calculates
the cardiac output of the patient by using the time the bolus takes to travel through the
hearts right side. The nurse inserts a thermodilution pulmonary artery catheter, also
known as the SwanGanz catheter, as shown in Figure 8.9, into the patient, usually at the
internal jugular vein or subclavian vein (in the neck area). This catheter has a small balloon that once inflated naturally floats toward the heart to the pulmonary artery.

20

Heart and Circulation

Figure 8.10 The relationship of the temperature gradient and time. (Adapted from Baker, P. D., Orr,
J., Westenskow, D. R. and Johnson, R. W. Method and apparatus for producing thermodilution
cardiac output measurements utilizing a neural network, US Patent, 5,579,778.)

The cardiac output can be calculated from (Webster, 1998)

CO =

|Q|
t1

bcb | Tb (t ) | dt

(m3/s)

(8.5)

where Q is the heat injected in joules, b is the density of the blood in kg/m3, cb is the
specific heat of the blood in J/(kgK), and Tb is the temperature gradient function as
shown in Figure 8.10. The area under the temperature versus time curve is inversely proportional to the blood flow or cardiac output. Therefore, healthy patients with normal
cardiac output will have less area under temperature versus time curve, than unhealthy
patients with some type of coronary insufficiency. If cardiac output is reduced, the bolus
takes longer to pass through the heart.
Example 8.2 Given the following data: blood density = 1060 kg/m3, heat capacity of the
blood = 3640 J/(kgK), amount of heat injected = 1.25 J, and the area under the curve in
Figure 8.10 is 1.0 sC, calculate the cardiac output.
From Eq. (8.5), the cardiac output can be computed as

8.4 Cardiac Output

CO =

1.25 J
3

1060 kg/m 3640 J/(kg C) (1.0 s C)

= 3 10 7 m 3 /s.

21
(8.6)

The disadvantage of this method is its invasiveness. Also, the lower temperature
bolus can be rapidly warmed to body temperature. Repeat measurements are easier to
perform than for the dye dilution technique (see below). It is more difficult to calculate
the size of the temperature perturbation than it is to measure the quantity of injected dye,
however. Another disadvantage of the thermodilution method is that the injecting catheter
acts as a heat exchanger between the blood and the cold saline. The cold saline left at the
end of the injection can further cool the blood. However, the error is not too significant.
8.4.3 Dye dilution

The principle of this technique is similar to that of thermodilution. A small quantity of

dye is injected into the venous blood stream through a catheter tip at a region very close
to the heart. The dye passes through the heart, and is mixed with blood. The dye then
appears in the arterial circulation. Arterial blood is drawn off through another catheter by
a motorized syringe and passes at a constant rate through an optical densitometer. A densitometer measures the concentration of the dye within the blood. The most commonly
used dye is indocyanine green, which has low toxicity and does not remain in the blood
stream for extended periods of time.
The disadvantages of the dye dilution technique are: it is not completely nontoxic and the dye cannot be removed immediately from the blood stream, thus repeat
measurements are difficult to perform (Cohn and Wayne, 1982).
8.4.4 Radionuclide angiography

External detectors are used to record activity from short-lived radiopharmaceuticals. This
is a procedure that yields a high degree of contrast between the blood and the surrounding
structures. There are two classes of radionuclides:
1. Positron emission tomography (PET), which uses substances that release positrons, which travel a short distance and then react with electrons to emit pairs of photons
on paths 180 apart. This technique is excellent for measuring the extent of myocardial
ischemia and is highly specific for the diagnosis of coronary artery disease (Marwick
1995).
2. Single photon emission computerized tomography (SPECT), which is a
method based on radionuclides such as 99mTc. The imaging device in SPECT is usually a
gamma-ray camera. Diagnostic accuracy and adjunctive assessment of ventricular function make SPECT the currently favored stress imaging radioisotope technique ( Merz
1997).
Chapter 7 discusses the principles of radionuclides.

22

Heart and Circulation

8.4.5 Echocardiography

Sound of frequencies greater than 20 kHz, the highest frequency that a human ear is able
to detect, are called ultrasound. The application of the ultrasonic imaging technique to the
heart is called echocardiography. This technique is noninvasive. The transducer contains
a piezoelectric crystal, which transmits and receives ultrasound resulting in images.
Echocardiography can measure the size of the heart, functions of the heart, the blood flow
through the heart (when combined with the Doppler technique), and the movement of the
heart wall. It can help identify tumors and clots in the heart. Structural abnormalities of
the heart wall, valves, and the blood vessels going in and out of the heart can be detected
by this technique as well.
Cardiac diagnosis uses high-frequency ultrasound (1.6 to 2.25 MHz). The sound
waves are both emitted and received by the same sensor. Sound waves are reflected at
interfaces between media of different acoustic impedances. Motion mode, or M-mode,
echocardiograms are generated by means of a single ultrasonic beam traversing cardiac
structures with definable anatomic shapes (chamber walls, interventricular septum) or
characteristic motions (cardiac valves). The echocardiograms are then recorded on an
oscilloscope, where the time-of-flight of reflected sound occupies the ordinate, and
slower time the abscissa. Advances in echocardiographic equipment including real-time
cross-sectional scanning (two-dimensional echocardiography), Doppler echocardiography,
and contrast echocardiography have broadened the role of echocardiography in the diagnosis of a variety of cardiac diseases.
M-mode echocardiography

With this method, the ultrasonic sensor is placed on the anterior chest. The beam passes
through the right ventricle, interventricular septum, left ventricular cavity, posterior left
ventricular wall, pericardium and lung. Figure 8.11 shows that these structures are portrayed on the ordinate and time is displayed on the abscissa. Additional information is
obtained by sweeping the sensor in an arc between the base and apex of the heart. By
proper positioning of the sensor all four cardiac valves, both ventricles, and the left
atrium can be visualized (Bom, 1977). M-mode echocardiography provides good axial
resolution and is therefore superior in qualitative diagnoses such as pericardial thickening,
small effusion, and constrictive hemodynamics (Cikes and Ernst, 1983). However, because it lacks spatial geometry, most potential information from using many cardiac
cross-sections is unavailable (Cikes and Ernst, 1983).

8.4 Cardiac Output

23

Figure 8.11 Simultaneous recording of motion mode (M-mode) and two dimensional echocardiograms. The arrows on the right image indicates the position of the ultrasound beam from which the
M-mode recording was made. LVW = left ventricular wall, LV = left ventricle, LA = left atrium,
RV = right ventricle (From College of Veterinary Medicine, Univ. of Tennessee. 2000. M-Mode
Echocardiography [Online] http://www.vet.utk.edu/m-mode.html)

Two-dimensional echocardiography (2-D echo)

M-mode echocardiography provides extremely useful information about cardiac structures, but cannot depict lateral motion (motion perpendicular to the ultrasonic beam). In
two-dimensional echocardiography, the ultrasonic beam angle is swept very rapidly and
can provide pictures of lateral motion and cardiac shape not available by M-mode echocardiography alone. Images recorded in 2-D echo are displayed on videotape. A variety
of views are possible but those commonly employed are the long axis view, the crosssectional view, and the four-chamber view (Figure 8.12). Two-dimensional echocardiography is superior to M-mode echocardiography in evaluating left ventricular function,
left ventricular aneurysm, and intracavitary thrombus. Table 8.2 shows the relative
advantages of M-mode and 2-D echocardiography.

24

Heart and Circulation

(a)

(b)

Figure 8.12 2-D echocardiography of the long axis view of the right ventrical (RV): (a) the ultrasonic beam angle through the heart, (b) the cross-sectional diagram of the image and (c) the actual
2-D image. TV = tricuspid valve, AML = anterior mitral leaflet. (Adapted from Rafferty, T. 1992.
Transesophageal two-dimensional echocardiography
http://www.gasnet.org/reference/echomanual/html/2-d_echo..html)

(c)

8.4 Cardiac Output

25

Table 8.2 Relative advantages of echocardiographic examination techniques. (Adapted from Roelandt, 1983)
M-mode echocardiography

Two-dimensional echocardiography

Excellent time resolution

Accurate dimensional measurements
Timing of events against other parameters
Easy storage and retrieval

Anatomical relationships
Shape information
Lateral vectors of motion
Easier to understand

Doppler echocardiography

Doppler echocardiography employs the Doppler principle, which is discussed in section

8.8.3. When the ultrasonic wave is reflected by a moving object, the frequency of the
reflected beam is shifted. Doppler echocardiography was originally employed to evaluate
blood flow in peripheral arteries and veins; however, Doppler is now combined with
standard cardiac ultrasound. Doppler studies can measure the transvalvular pressure gradient in patients with aortic, pulmonic, tricuspid, and mitral stenosis. There is a good correlation between the gradient measured by Doppler and that obtained with catheterization.
Contrast echocardiography

Contrast echocardiography is based on the fact that the injection of almost any liquid
through a small catheter produces a cloud of echoes on the M-mode echocardiogram. The
most commonly used dye for contrast echocardiography is indocyanine green. The microbubbles are too large to pass through the capillaries and yield a cloud of echos. Thus,
an injection of indocyanine green on the right side of the heart remains in the right side of
the heart unless there is a right-to-left shunt. We measure the cardiac output by tracing
the intravascular bubbles with a high degree of sensitivity by ultrasound (Meltzer and
Roelandt, 1982).
8.4.6 Magnetic resonance imaging (MRI)

MRI provides higher spatial resolution than ultrasound and the capability of measuring
the velocities in the three spatial directions for any plane through the heart (Ingels et al.,
1996). In contrast to radionuclide angiography, MRI utilizes no ionizing radiation. It can
also diagnose the characteristics of the cardiac muscle. Also, the stroke volume can be
calculated by integrating the flow curve in the main pulmonary artery during a cardiac
cycle. MRI offers excellent anatomical detail of the cardiovascular system, particularly in
congenial heart disease and in diseases of the aorta (Underwood, 1992).

8.5 Cardiac Sounds

26

Heart and Circulation

Cardiac sounds shown in Figure 8.3 are associated with the movement of blood during
the cardiac cycle. Table 8.3 gives descriptions of the origins of various heart sounds. The
first heart sound is low-pitched, soft, and relatively long. It is associated with the close of
the AV valves. The opening of the valves does not produce any sound. The sounds are
caused by vibrations within the walls of the ventricles and major arteries during valve
closure, not by the valves snapping shut. The first heart sound signals the onset of systole.
The second heart sound has a higher pitch, is shorter and sharper. It is associated with the
closure of the semilunar valves that occurs at the onset of ventricular relaxation as the left
and right ventricular pressures fall below the aortic and pulmonary artery pressures, respectively. Thus the second heart sound signals the onset of ventricular diastole. Murmurs are vibrations caused by turbulence in the blood moving rapidly through the heart.
Table 8.4 lists the types of murmurs and what they represent.
Table 8.3 The heart sounds. The 1st and 2nd heart sounds are most prominent.
Sound

Origin

1st sound
2nd sound
3rd sound
4th sound

Closure of mitral and tricuspid valves

Closure of aortic and pulmonary valves
Rapid ventricular filling in early diastole
Ventricular filling due to atrial contraction

Table 8.4 Timing of murmurs. For example, if the physician hears the 1st heart sound, a swishing
sound, and then the 2nd heart sound, the patient likely suffers from AV valve insufficiency.
Characteristic

Type of murmur

Valve disorder

Systolic murmur

Whistling
Swishing

Stenotic semilunar valve

Insufficient AV valve

Whistling
Swishing

Stenotic AV valve
Insufficient semilunar valve

1st HSmurmur2nd HS
Diastolic murmur
2nd HSmurmur1st HS

A stenotic valve is a stiff, narrowed valve that does not open completely. Thus,
blood must be forced through the constricted opening at tremendous velocity, resulting in
turbulence that produces an abnormal whistling sound similar to the sound produced
when you force air rapidly through narrowed lips to whistle. An insufficient valve is one
that cannot close completely, usually because the valve edges are scarred and do not fit
together properly. Turbulence is produced when blood flows backward through the insufficient valve and collides with blood moving in the opposite direction, creating a swishing or gurgling murmur (Sherwood, 2001). Improved diagnosis can be obtained from
heart sound audiograms, where the magnitude of each frequency component is displayed.
8.5.1 Stethoscopes

8.5 Cardiac Sounds

27

A stethoscope is used to transmit heart sounds from the chest wall to the human ear.
High-frequency sounds, or murmurs, are easier to hear with the diaphragm. The bell,
which should be applied lightly to the chest, transmits low-frequency sounds more effectivelyfor example, the diastolic murmur of mitral stenosis and third and fourth heart
sounds. Murmurs arise from turbulent blood flow and are characterized by their timing,
quality, and intensity. Intensity is graded from 1 (just audible) to 6 (audible without a
stethoscope).

Figure 8.13 A stethoscope with bell and diaphragm modes. Adapted from Mohrin, C. M., 1995.
Stethoscope. US Patent, 5,389,747.

Figure 8.13 shows a diagram of a stethoscope which incorporates both the bell
and diaphragm modes into the same chestpiece. Using this particular model (Mohrin,
1995), we can interchange between the bell and diaphragm modes by simply pressing the
chestpiece against a patients body and twisting the bell housing so that the two dia-

28

Heart and Circulation

phragms (shown in Figure 8.13) adjust their openings. If the two openings of the diaphragms coincide with each other, the stethoscope is operating in the bell mode. If there
is no through opening, it is in the diaphragm mode. Thus, physicians are able to change
between the bell and diaphragm modes without having to remove the bell housing from
the patients body.
8.5.2 Microphones

Microphones in use today are either crystal microphones (piezoelectric effect) or dynamic microphones (Faradays principle) (Peura and Webster, 1998). Piezoelectric sensors are used as microphones since they are in the form of a thin film. They are very useful if one is interested in detecting surface vibrations of an object, such as the heart sound.
When a force, f, is applied to a polarized crystal, the resulting mechanical deformation
gives rise to an electric charge Q. Microphones turn an acoustical pressure into a voltage.
Q = kf = Kx

(8.7)

where k is the piezoelectric constant in C/N, K is the proportionality constant in C/m, and
x is the deflecting distance of the piezoelectric sensor.
Figure 8.14 illustrates a charge amplifier. Chapter 2 gives the voltage across an
initially uncharged capacitor, C, by
v=

1
C

t1

idt =

1
C

t1

dx
x
dt = K
C
dt

(8.8)

where t1 is the integrating time and i is the current flowing through the capacitor, C. A
charge amplifier acts as a high-pass filter, and Figure 8.15 shows the time constant is

= RC
Hence it only passes frequencies higher than the corner frequency fc = 1/(2RC).

(8.9)

8.5 Cardiac Sounds

29

Figure 8.14 The piezoelectric sensor generates charge, which is transferred to the capacitor, C, by
the charge amplifier. Feedback resistor R causes the capacitor voltage to decay to zero.

Figure 8.15 The charge amplifier responds to a step input with an output that decays to zero with a
time constant = RC.

30

Heart and Circulation

8.6 Myocardial Viability

The assessment of myocardial viability is one of the newest and most rapidly expanding
areas of clinical and experimental research in cardiology. A number of diagnostic techniques have been developed to help the clinician determine with a good degree of accuracy whether myocardial viability is present or absent in dyssynergic (failure of muscle
coordination) regions of the heart.
Several methods have been used to assess the presence or absence of myocardial
viability. These methods depend on the demonstration of myocardial perfusion (presence
of oxygen and nutrients), myocardial metabolism, or a response to a stimulus. The methods that use the demonstration of perfusion as the basis for the presence of viability include positron emission tomography (PET), thallium-201 imaging, and myocardial contrast echocardiography.
8.6.1 PET

Positron emission tomography uses isotopes that emit photons 180 apart (i.e. in opposite
directions). It has been successfully used to assess myocardial viability. This technique
has the distinct advantage of allowing for the assessment of both myocardial perfusion
and metabolism. This is particularly important since severely ischemic but viable myocardium might show almost no flow, despite the presence of myocardial metabolism.
Under normal circumstances, the myocardium uses fatty acids for its metabolism. However, when blood flow slows and oxygen supply decreases, anaerobic metabolism develops, and consequently, the myocardium begins using glucose instead of fatty acids as a
source of energy. A high ratio of regional glucose utilization to myocardial flow has been
found to be a reliable sign of severely ischemic but viable muscle.
Several studies have demonstrated that PET is a better predictor of myocardial
viability than thallium-201 scintigraphy. This probably relates to the improved imaging
resolution of PET and to its added benefit of also assessing metabolism. The advantage of
PET over thallium is lost when sophisticated quantitative analysis of perfusion by thallium is undertaken. The disadvantages of PET are its high cost and nonportability.
8.6.2 Thallium-201 imaging

Thallium, a member of the heavy metal group of elements, had little biologic use other
than as a rodenticide until its application in myocardial imaging in the early 1970s. Although a heavy metal capable of causing typical heavy metal poisoning at high doses,
doses of 5,000 to 10,000 times less than that associated with clinical poisoning can be
employed for myocardial imaging. The biologic properties of thallium at these low concentrations closely resemble those of potassium.
Thallium-201 imaging of the myocardium is one of the more difficult nuclear
medicine procedures from the standpoint of technique. The relatively low photon abundance and low energy of the imaged photons require excellent equipment and technique

8.6 Myocardial Viability

31

to obtain high-quality images. The cost of Thallium-201 and the time required for exercise testing and imaging makes the repetition of technically inadequate studies difficult
(Ritchie et al., 1978).
8.6.3 Myocardial contrast echocardiography

The principle of this technique is discussed in section 8.4.5.

8.7 Circulation
The three kinds of blood vessels are arteries, veins, and capillaries, which in the human
body have been estimated to extend a total distance of 100,000 km. Arteries carry blood
away from the heart to organs throughout the body. Within these organs, arteries branch
into arterioles, tiny vessels that give rise to the capillaries. Capillaries form networks of
microscopic vessels that infiltrate each tissue. It is across the thin walls of capillaries that
chemicals are exchanged between the blood and the interstitial fluid surrounding the cells.
At the downstream end, capillaries rejoin to form venules, and these small vessels converge into veins. Veins return blood to the heart. Notice that arteries and veins are distinguished by the direction in which they carry blood, not by the quality of the blood they
contain. Not all arteries carry oxygenated blood, and not all veins carry blood depleted of
oxygen. However, all arteries do carry blood from the heart to capillaries, and only veins
return blood to the heart from capillaries. The arterial walls are muscular, thus the arteries
are able to control the blood flow through constriction and dilation. The size of the capillary is 10 m in diameter. The capillaries are where the gaseous exchange of CO2 and O2
occurs.
Peripheral vascular diseases are a group of distinct diseases or syndromes involving the arterial, venous, or lymphatic system (Strandness et al., 1986). The major
vascular disease is coronary artery disease, which kills more than a million people every
year, and afflicts more than 5 million others with disabling symptoms. The development
of various noninvasive tests for diagnosis of arterial disease continues to improve rapidly.
These tests enable us to diagnose and follow the course of arteriosclerosis (abnormal
condition of an artery) much more accurately than was possible in the past. The Doppler
ultrasonic method is the most reliable method of evaluating the severity of lower extremity ischemia and monitoring postoperative results. A promising new development is Duplex scanning (section 8.11.1) of the peripheral arteries, combining Doppler spectral
analysis of blood flow with B-mode ultrasound imaging. Using the Duplex scan to localize the stenosis and to accurately estimate its severity may allow selected vascular reconstructions to be done in the future without preoperative arteriography.
Venography remains the standard for the diagnosis of deep venous thrombosis
(blood clots), but does have significant limitations. In addition to the cost and discomfort,
some venograms are not diagnostic because of technical reasons. Likewise, invasive
techniques have become useful in the long-term evaluation of treated and untreated peripheral vascular disease. The increasing sophistication of instrumentation has provided

32

Heart and Circulation

not only physiological information but also anatomic depiction of intraluminal lesions by
ultrasonic imaging. Direct arteriography has also seen new developments with the advent
of digital subtraction angiography. While venous digital subtraction angiography has not
proven to be as clinically useful as originally hoped, application of digital subtraction
techniques to direct arteriography has permitted significant reduction in amounts of contrast agent required. By combining digital subtraction techniques with direct arterial angiography, patients at high risk of renal failure can be studied safely while combined arterial systems such as abdominal and carotid vessels can be simultaneously studied in patients with normal renal function. As for the future, magnetic resonance imaging (MRI)
holds promise as an arterial diagnostic tool, particularly with large vessels. However, it
appears unlikely that it will replace ultrasound and contrast arteriography.

8.8 Blood Flow

Physical factors that influence blood flow are pressure and resistance. The flow through
arterial grafts is measured at the time of surgery to ensure that the graft has been successfully inserted. The flow in peripheral blood vessels is measured as an aid in the diagnosis
of peripheral vascular disease.
The flow rate F of a single vessel is the volume of blood moving past a fixed
point per unit time. F can be calculated from
F = P/R, (mL/min).

(8.10)

Pressure, P, is usually measured in mmHg and R is the resistance.

R= 8L/r4,

(8.11)

where L = length, = viscosity, r = radius. Thus,

F = Pr4/(8L) (Poiseuilles law).

(8.12)

P is the difference between the mean arterial blood pressure (MABP) and right atrial
pressure. Since the right atrial pressure is about zero, P = MABP. R is the resistance
through all vascular beds in parallel. Fung (1997) provides further information on the
biomechanics of the blood vessels.
For detecting or measuring blood flow, ultrasonic Doppler flowmetry has become the main technique of noninvasive investigation, both in the clinical and research
laboratory. The continuous wave (CW) and pulsed varieties each have their advocates,
the former being easier to use while the latter affords more precise flow interrogation.
Although the audible signal suffices for many clinical applications, recordings are required for others and for most research endeavors. Recordings made with the zerocrossing detector are adequate to some extent for real-time frequency spectrum analyzers.
These devices not only depict the velocity flow envelope with precision but also allow
analysis of the power spectrum at each frequency. Increased velocity and disturbances in

8.8 Blood Flow

33

the pattern of blood flow, revealed by these processing methods, are good indicators of
vascular disease. Measurement of volume flow requires knowledge of the cross-sectional
area of the vessel (which can be obtained from the B-mode image) and the mean of the
velocity profile. Since velocity profiles are seldom completely axisymmetric, precise
flow measurement has been difficult to achieve with either the CW or pulsed-Doppler.
Laser Doppler methods are applicable only to the cutaneous tissues and are difficult to quantify. At present their role remains uncertain. The use of magnetic resonance
imaging (MRI) to measure blood flow, though promising, is in its infancy. Positron emission tomography (PET) has proved useful in the evaluation of local blood flow, especially in the brain, but is not widely available and remains prohibitively expensive. The
transcutaneous electromagetic flowmeter seems to have little merit. Monoitoring the
clearance of radionuclides, a semi-invasive technique for evaluting blood flow in isolated
areas, was initially greeted with enthusiasm; however, the washout curves proved to be
difficult to interpret, and the results were inconsistent. At present isotope clearance is
employed only for special clinical studies or for research work.
8.8.1 Dilution

This technique was discussed in section 8.4.3. Applying a similar approach, we can determine the blood flow.
8.8.2 Electromagnetic flowmeter

This technique utilizes Faradays law of magnetic induction, which states that a voltage is
induced in a conductor that is moving in a magnetic field. It can measure pulsatile flow.
The blood is an electrical conductor, therefore voltage will be developed when the blood
flows through a magnetic field. Figure 8.16 illustrates the principle of an electromagnetic
flowmeter. If the magnetic field, the direction of motion and the induced voltage are mutually at right angles (orthogonal), then
e = Blu

(8.13)

The induced voltage e is proportional to the velocity of the conductor u (averaged over the vessel lumen and walls), the strength of the magnetic flux density B, and
the length between electrodes, l.

34

Heart and Circulation

B
l

S
e

Figure 8.16 Principle of an electromagnetic flowmeter.

Small-size probes can be made using an iron-cored electromagnet to increase the

signal strength. Air-core coils can be used for the larger probes to reduce the problems of
hysteresis in the magnetic cores. The electrodes make contact with the outer wall of the
blood vessel.
Dc excitation

If the generation of the magnetic field is done with dc, the induced voltage at the electrodes is also dc. The induced voltage is small compared to surface potentials generated
at the electrodes. It is difficult to distinguish changes in blood flow from changes in the
surface potentials due to polarization at the electrodes and electrodes may detect other
signals such as electric impulses from the heart. Thus dc excitation is not used commercially.
Ac excitation

If the generation of the magnetic field is done with ac then the induced voltage at the
electrodes is alternating. The surface potentials at the electrodes are removed by capacitive coupling. The ac excitation causes a different problem than dcthe change in magnetic field intensity causes an induced current in the conductor, which is proportional to
the rate of change of the magnetic field intensity. The induced current cannot be separated from the desired signal at the electrodes. One of the solutions is to measure the
voltage from the electrodes when the magnetic field intensity is not changing.

8.8 Blood Flow

35

The electromagnetic flowmeter measures the average of the blood velocity

across the lumen of the vessel. The outside diameter of the vessel is constrained to be the
same as the internal diameter of the probe, so the volume flow through it can be calculated. In practice, this is not an ideal assumption. Most blood vessels display laminar
flowflow is greatest in the center of the vessel and slowest near the walls. It is not uniformly sensitive to the same blood velocity at different positions across the lumen of the
vessel. Alterations in the velocity profile across the vessel therefore alter the measured
mean velocity, giving rise to error. Fortunately, this induced error is not large enough to
cause unacceptable measurements.
8.8.3 Ultrasonic flowmeter

Ultrasonic flowmetry is a commonly used technique for measuring blood velocity in the
peripheral arteries. It can measure pulsatile flow. A piezoelectric sensor is used to convert
from electric to acoustic signals (see section 8.5.2). Figure 8.17 shows a system for blood
flow measurement using an ultrasonic flowmeter. The Doppler effect shifts the frequency
of the ultrasonic beam from the oscillator when it intercepts the moving blood cells by an
amount f, which can be computed from
f =

2uf 0 cos
,
c

(8.14)

where f0 is the fundamental frequency of an ultrasonic wave from the source, traveling at
velocity c through the blood. The ultrasonic wave intercepts a stream of moving blood
with velocity u crossing the ultrasound beam at an angle to produce the frequency shift
f. The factor of 2 occurs because the Doppler shift arises both on absorption of the
sound by the moving blood particles and between the transmitting blood cell and the receiving transducer.

36

Heart and Circulation

Oscillator

Scan head
Image plane

Skin surface

Blood vessel
Doppler
angle
Computer

RF
amplifier

Detector

AF
amplifier

F/V
converter

Figure 8.17 Ultrasonic flowmeter. The sensor at the scan head transmits the signal from the oscillator and receives the reflected wave from the blood cells. The RF (radio frequency) amplifier amplifies the received signal and the carrier frequency, then AF (audio frequency) signal is produced by
a detector. Adapted from Picot, P. A. and Fenster, A. 1996. Ultrasonic blood volume flow rate meter. US Patent, 5,505,204.

The Doppler probe has a piezoelectric ultrasound transducer integrated into the
tip. The crystal on the tip sends and receives ultrasound waves. The timing of the sending
and receiving velocity allows the ultrasonic flowmeter to measure blood flow velocities.
The returning signal is transmitted, in real time, to the display device. The display shows
the average velocity of all the red cells within the sample volume. When the sample area
is constant, the peak velocity is accurately tracked in the center of the artery and the key
parameters remain relatively positionally insensitive and reliable.
Computerized parameters of intracoronary flow velocity, including peak and
mean diastolic and systolic velocities, diastolic and systolic velocity integrals, mean total
velocities, and the total velocity integral are automatically analyzed.
8.8.4 Laser Doppler flowmeter

The Doppler ultrasonic flowmeter can assess blood flow from the Doppler shift that
sound waves experience when they travel through a large blood vessel. However, the
ultrasonic flowmeter is not able to measure the blood flow in small blood vessels, or
microcirculation. In contrast, this is possible if laser light is used instead of sound (Shep-

8.8 Blood Flow

37

herd and berg, 1990). Light is capable of measuring the velocities of red blood cells
even at the relatively slow speeds with which they move through capillaries. LaserDoppler blood flowmetry thus ulilizes the short wavelengths (extremely high frequencies)
of visible and infrared light to measure the blood flow velocity in a similar manner as the
Doppler ultrasound blood flowmeter. The spectral purity of the laser makes it practical to
detect the slight frequency shifts produced by the interactions between photons and moving red blood cells.
Figure 8.18 shows the principle of Laser-Doppler measurement. Laser light is
transmitted to the blood vessels and reflected to a detector. Light that intercepts the moving red blood cells is Doppler-shifted. Light in tissue is scattered by stationary tissue and
it reaches the detector without experienceing a Doppler shift (Shepherd and berg, 1990).
A laser-Doppler catheter velocimeter can be used to measure intravascular blood
flow velocity. This can be achieved easily because optical fibers have minute diameters
and permit high resolution. However, catheters disturb the flow of blood in their
neighborhood, and because light is multiply scattered in passing through the blood, combining the high spatial and temporal resolution of laser-Doppler velocimetry with catheter
delivery is not straightforward. At the present time, the development of instruments for
the practical measurement of blood flow velocity through a catheter is an area of active
research.
The major advantage of the laser-Doppler flowmeter is the ability to measure
blood flow in regions other than the limbs. Another advantage is its frequency response.
Not only can the laser-Doppler flowmeter be used to examine variations in blood flow
through the cardiac cycle, but also, on a somewhat longer time scale, slower rhythms can
be observed.

Figure 8.18 Laser-Doppler flowmetry. Light that intercepts the red blood cells experiences a Doppler frequency shift.

8.9 Blood Pressure

38

Heart and Circulation

Arterial pressure is an essential check on the performance of the heart. The systolic and
diastolic pressures not only provide those maximum and minimum values that are an important aspect of the pulsations, but also permit a rough estimate of mean pressure in
large arteries. Normal blood pressure is considered to be 120/80 mmHg. Mean blood
flow is essentially the same at all cross sections of the arterial tree, in the total population
of peripheral arterioles as in the ascending aorta. Mean pressure, however, falls continually from the ascending aorta to the terminus of the vena cava, an indication that the energy imparted to the blood by ventricular ejection is gradually dissipated as it travels
through the vascular system. The decrement in pressure per unit length of the system, or
vascular resistance, is greatest in the microcirculation.
Blood pressures measured in the distal portions of the extremities are especially
useful for identifying the presence of hemodynamically significant disease. The degree of
circulatory impairment is usually reflected by the relative or absolute reduction in resting
pressure. Stress testing increases the sensitivity of pressure measurement, permitting the
recognition of clinically significant disease in limbs in which resting pressures are
borderline normal. Unfortunately, pressure measurements are not always reliable.
8.9.1 Indirect measurement

Figure 8.19 shows a sphygmomanometer, which indirectly measures blood pressure. An

inflatable cuff formed by cloth containing an inflatable bladder encircles the arm. A hand
or electric pump and a pressure gage are used. A pressure release valve allows controlled
release of pressure within the cuff.

8.9 Blood Pressure

39

Figure 8.19 The sphygmomanometer detects arterial opening and closing that occurs between systolic and diastolic pressures.

The auscultatory method of measuring blood pressure refers to listening to

sounds produced by blood turbulence within the artery. The occlusive cuff is pumped up
to about 180 mmHg, then bled off at about 3 mmHg/s. When the arterial pressure exceeds
the cuff pressure, blood squirts through the partially occluded artery and creates turbulence, which creates Korotkoff sounds (Figure 8.20). A stethoscope placed distal to the
cuff over the brachial artery detects the tapping noise that signals systolic pressure at
about 120 mmHg. When the cuff pressure decreases below about 80 mmHg, the artery
remains open the entire cycle. The sounds disappear, which indicates diastolic pressure.
A piezoelectric sensor placed within the cuff replaces the stethoscope for automatic detection.
Cuff

Occluded
blood vessel

Figure 8.20 The pressure of the cuff occludes the blood vessel. When the arterial pressure is
greater than the pressure applied by the cuff, Korotkoff sounds are created and blood pressure can
be measured.

40

Heart and Circulation

There are several automatic noninvasive blood pressure monitoring devices

available commercially. A typical automatic noninvasive blood pressure monitor is composed of an inflatable cuff with a microphone. The signal from the microphone is then
passed to a filter. Generally, the cuff method gives systolic pressures within 2 mmHg of
the values obtained by direct arterial puncture and diastolic pressure within 4 mmHg.
Pressure in the cuff is sometimes measured with a mercury manometer, eliminating the
need for calibration.
The oscillometric method uses the same cuff. When the artery changes blood
volume under the cuff, this changes air volume within the cuff, which changes pressure
within the cuff. These pressure fluctuations can be seen when using an aneroid or mercury manometer and the cuff pressure is between systolic and diastolic. Alternatively a
pressure sensor (piezoelectric sensor) can detect these fluctuations in an automatic instrument. Because the pressure sensor is within the instrument, no separate sensor is
within the cuff and only a single air tube leads to the cuff.
Figure 8.21 shows the indirect measurement of blood pressure. So is the point
where cuff-pressure oscillations start to increase. As is the amplitude corresponding to
auscultatory systolic pressure and Ad is the amplitude corresponding to auscultatory diastolic pressure. Am is the maximal oscillation amplitude, which signals mean pressure.

Figure 8.21 Top: Cuff pressure with superimposed Korotkoff sounds, which appear between systolic and diastolic pressures. Bottom: the oscillometric method detects when the amplified cuff
pressure pulsations exceed about 30% of maximal pulsations. From Geddes, L. A. Cardiovascular
devices and their applications. New York: Wiley. 1984.

8.9.2 Direct measurement

8.9 Blood Pressure

41

Arterial and venous blood pressure can be measured by inserting a catheter into the blood
vessel and maneuvering it until the end is at the site at which the blood pressure is to be
measured. In this case the diaphragm of the sensor is mounted at the end of the catheter
outside the body (similar to section 8.2). Catheters for use in the artery are usually thin,
short, and highly flexible. A catheter can be inserted into the artery inside a needle as
well. The alternative method is to use a catheter-tip sensor where the miniature diaphragm is at the tip inside the vessel. Although accurate, these methods are very invasive.
8.9.3 Arterial tonometry

Arterial tonometry is a noninvasive technique for monitoring the arterial blood pressure
in a continuous manner. A linear array of pressure sensors is pressed against the radial
artery so that at least one sensor is directly over the lumen. Pressure is increased from
low to high so that measurements are made when the artery is half collapsed. By selecting
the maximal peak-to-peak pressure reading from all sensors, the arterial pressure is determined in the same way as for the applanation tomometer (section 7.10.4). Zorn et al.
(1997) compared the Colin Pilot 9200 tonometric blood pressure measurements with intra-arterial blood pressure measurements. Tonometric values were slightly less than the
intra-arterial pressure measurements; the mean difference for systolic blood pressure was
2.24 8.7 mmHg and for diastolic pressure was 0.26 8.88 mmHg.

8.10 Vessel Distension

Blood vessels can be distended (stretched) by raising intravascular pressure or decreased
in radius by lowering it since they are elastic. The difference between the pressures inside
and outside the vesselthe transmural pressureis one of the factors that controls the
radius. The contraction or relaxation of vascular smooth muscle alters vessel diameter by
changing the elasticity of the wall and the elasticity of a vascular bed determines how
much of the blood volume is accommodated within that region at the existing local pressure.
The distribution of the blood volume among different parts of the system is determined by the relation between local pressure and vascular distensibility. Pressure is
high in arteries, but they are not very distensible and the volume of blood in the arterial
tree is relatively small. In contrast, a large part of the blood volume resides in the veins,
even though their pressure is low, because they are readily distended. The distension of
an arterial wall can be calculated from

Distension (%) =

d
100
dD

(8.15)

where d is the diameter change and dD is the diastolic diameter (Nakatani et al., 1995).

42

Heart and Circulation

8.10.1 Intravascular ultrasound

Intravascular ultrasound provides both in vitro and in vivo two-dimensional visualization

of arteries in real time. Figure 8.22 shows that a rotating ultrasonic transducer illuminates
the walls. Reflections from the inner and outer wall surfaces yield accurate determination
of artery luminal dimension, cross-sectional area, wall thickness, and wall morphology
(Forestieri and Spratt, 1995).
Ultrasonic
beam
Rotating
Ultrasonic
catheter
transducer

Blood

Arterial
wall

Figure 8.22 A catheter is inserted through the blood vessels. A rotating ultrasonic transducer is
attached at its tip and illuminates the walls.

8.10.2 Angiography

Peripheral angiograms are most commonly done to test the arteries that supply blood to
the head and neck or the abdomen and legs. Because arteries do not show up well on ordinary X rays, angiograms utilize a contrast agent containing iodine, which is injected
into the arteries to make them visible on radiographs. A catheter is placed into the artery,
via the groin, and is manipulated by the physician into the artery requiring study. Once
the catheter is in place, contrast agent, or dye is injected through the catheter into the arteries and a series of X rays taken. The dye allows the doctor to see the inside of the patients artery and determine how well blood is moving through the vessel.
In digital subtraction angiography (DSA), a first image is taken without contrast
agent. Then a second image is taken with contrast agent. The two images are digitized,
then subtracted, yielding no image in all regions without contrast agent. Vessels containing constrast agent show very well, so less contrast agent is required.

8.11 Vessel Volume Flow

8.11 Vessel Volume Flow

43

One application of vessel volume flow measurement is the prediction of stenosis of both
common and internal carotid arteries by monitoring common carotid blood flow. Generally, an adequate degree of stenosis (usually accepted as 50% diameter reduction), results
in a measurably decreased volume flow. Vessel volume flow measurements may also be
applied in the diagnosis and treatment of vascular malformations. Specifically, the measurement of volume flow may help to distinguish arteriovenous malformations and fistulae,
which are high-flow lesions, from venous malformations, which are low-flow lesions.
Moreover, volume flow measurements provide a quantitative way both of assessing blood
steals (deviation of normal flow) and of evaluating the effectiveness of embolization
(mass of foreign matter in the blood vessel) therapy. Renal dialysis patients may also
benefit from these measurements.
8.11.1 Flow visualization

Continuous-wave or pulsed-Doppler flow-mapping devices produces a static image on an

oscilloscope screen that corresponds to a projection of the lumen of the vessel underlying
the probe. The frequency of the Doppler signal is color-coded to indicate regions of increased velocity (stenotic areas); and, provided a pulsed-Doppler is used, the image can
be processed by a microcomputer to depict the outline of the vessel in a plane or produce
a histogram of its cross-sectional area. Although B-mode ultrasonic scanning allows large
vessels, such as the abdominal aorta, to be visualized, it is inappropriate for the routine
study of smaller vessels. With the development of real-time B-mode scanners, it is now
practical to examine virtually all vessels of interest, even those with diameters as small as
2.0 mm.
A major advance in flow visualization is the combination of B-mode scanning
and the pulsed-Doppler velocity detector in a single instrument, called the Duplex scanner. The two technologies are complementary. Doppler flow patterns help distinguish
between arteries and veins, and the image permits more accurate positioning of the Doppler sample volume. One instrument superimposes a real-time, color-coded Doppler
flow-map on a real-time B-mode image, thus incorporating the advantages of both modalities.
By using a gamma camera to trace the passage of an isotope through the veins of
the extremities, one can obtain crude, but sometimes clinically revealing flow-maps.
8.11.2 Ultrasonic sonography

To estimate the volume flow rate of blood through an artery, pulsed Doppler ultrasound
measures the velocity of the blood at 16 or more discrete locations across the vessel diameter. From the velocity measurement, and a measurement of the diameter of the vessel,
we estimate the volume of blood flowing through the vessel. Other techniques, such as
color M-mode, directly measure the one-dimensional velocity profile and are also used.
A clinical color Doppler ultrasound instrument is used with a position and orientation sensing device, and a computer with video digitizer to acquire blood velocity
measurements in two dimensions across an entire blood vessel lumen. Subsequently, the

44

Heart and Circulation

volume flow rate can be determined by integrating the two-dimensional velocity profile
over the vessel lumen area. The volume flow rate measurements are then displayed to an
operator in real time as a scrolling graph.
8.11.3 Video microscopy

With the developments in video and in computer technology over the past two decades,
we are now able to see structures that are very fine, movements that were too fast too be
seen, and images that were too dim or too noisy to be seen. Generally, the components
used in video microscopy are a camera, mixer, recorder, processor, analyzer, and monitor.
It also utilizes advances in image-processing (Inou, 1986). Chapter 6 discusses the principles of video microscopy.

8.12 References
Berne, R. and Levy, M. 1981. Cardiovascular Physiology. 4th ed. St. Louis: Mosby.
Bom, N. (ed.) 1977. Echocardiography: With Doppler Applications and Real Time Imaging. The Hague: Martin Nijhoff Medical Division.
Brown, B. H., Smallwood, R. H., Barber, D. C., Lawford, P. V., and Hose D. R. 1999.
Medical Physics and Biomedcial Engineering. Bristol, UK: IOP Publishing.
Burch, G. and Winsor, T. 1972. A Primer of Electrocardiography. 6th ed. Philadelphia:
Lea and Febiger.
Cikes, I. and Ernst, A. 1983. New aspects of echocardiography for the diagnosis and
treatment of pericardial disease. In J. Roelandt (ed.) The Practice of M-mode and
Two-dimensional Echocardiography. Boston: Martinus Nijhof.
Cohn, P. F and Wayne, J. 1982. Diagnostic Methods in Clinical Cardiology. Boston: Little, Brown.
College of Veterinary Medicine, University of Tennessee. 2000. M-Mode Echocardiography [Online] http://www.vet.utk.edu/m-mode.html
Davies, M. J., Anderson, R. H. and Becker, A. E. 1983. The Conduction System of the
Heart. London: Butterworth.
Forestieri, S. F. and Spratt, R. S. 1995. Angiography using ultrasound. US Patent,
5,394,874.
Fung, Y. C. 1997. Biomechanics: Circulation. 2nd ed. New York: Springer-Verlag.
Grossman, W. and Baim, D. S. 1991. Cardiac Catheterization. Angiography and Intervention, 4th ed. Philadelphia: Lea & Febiger.
Huang, S. K. S., and Wilber, D. J. (eds.) 2000. Radiofrequency Catheter Ablation of Cardiac Arrhythmias: Basic Concepts and Clinical Applications, 2nd ed. Armonk, NY:
Futura.
Ingels, N. B. Jr., Daughters, G. T., Baan, J., Covell, J. W., Reneman, R. S. and Yin, F. C.
1996. Systolic and Diastolic Function of the Heart. Amsterdam: IOS Press.
Inou, S. 1986. Video Microscopy. New York: Plenum Press.

8.12 References

45

Marwick, T. H., Shan, K., Go, R. T., MacIntyre, W. J. and Lauer, M. S. 1995. Use of
positron emission tomography for prediction of perioperative and late cardiac events
before vascular surgery. Amer. Heart J. 130 (6): 1196-202.
Meltzer, R. S. and Roelandt, J. (eds.) 1982. Contrast Echocardiography. Hague: Martinus Nijhoff.
Merz, C. N. and Berman, P. S. 1997. Imaging techniques for coronary artery disease:
current status and future directions. Clinical Cardiology. 20 (6): 526-32.
Mohrin, C. M. 1995. Stethoscope. US Patent, 5,389,747.
Moran, P. R. 1982. A flow zeugmatographic interlace for NMR imaging in humans.
Magn. Reson. Imaging 1: 197203.
Nakatani, S., Yamagishi, M., Tamai, J., Goto, Y., Umeno, T., Kawaguchi, A., Yutani, C.
and Miyatake, K. 1995. Assessment of coronary artery distensibility by intravascular
ultrasound: application of simultaneous measurements of luminal area and pressure.
Circulation 91: 290410.
Neuman, M. R. 1998a. Biopotential electrodes. In J. G. Webster (ed.) Medical Instrumentation: Application and Design. 3rd ed. New York: John Wiley & Sons.
Neuman, M. R. 1998b. Biopotential amplifiers. In J. G. Webster (ed.) Medical Instrumentation: Application and Design. 3rd ed. New York: John Wiley & Sons.
Onoda, M. 1994. Electrocardiograph system. US Patent, 5,284,151.
ORourke, M. 1995. Mechanical principles in arterial disease. Hypertension 26: 29
Orth, J. L. 1995. System for calculating compliance and cardiac hemodynamic parameters. US Patent, 5,423,323.
Peura, R. A. and Webster, J. G. 1998. Basic sensors and principles. In J. G. Webster (ed.)
Medical Instrumentation: Application and Design. 3rd ed. New York: John Wiley &
Sons.
Picot, P. A. and Fenster, A. 1996. Ultrasonic blood volume flow rate meter. US Patent,
5,505,204.
Rafferty, T. 1992. Transesophageal Two-Dimensional Echocardiography [Online]
http://www.gasnet.org/reference/echomanual/html/2-d_echo..html
Reinhard, C. J. and Fowler, K. A. 1987. Telemetry system and method for transmission
of ECG signals with heart pacer signals and loose lead detection. US patent,
4,658,831.
Ritchie, J. L., Hamilton, G. W. and Wackers, F. J. T. (eds.) 1978. Thallium201 Myocardial Imaging. New York: Raven Press.
Shepherd, A. P. and berg, P. . (eds.), 1990. Laser-Doppler Blood Flowmetry. Boston:
Kluwer Academic Publishers.
Sherwood, L. 2001. Human Physiology: From Cells to Systems. 4th ed. Pacific Grove,
CA: Brooks/Cole.
Strandness, D. E., Didisheim, P., Clowes, A. W. and Watson, J. T. (eds.) 1986. Vascular
Diseases: Current Research and Clinical Applications. Orlando: Grune & Stratton.
Tobuchi, K., Kato, T. and Namba, K., 1981. Silver-silver chloride electrode. US Patent,
4,270,543.
Togawa, T., Tamura, T. and berg, P. . 1997. Biomedical Transducers and Instruments.
Boca Raton FL: CRC Press.
Tompkins, W. J. (ed.) 1993. Biomedical Digital Signal Processing. Englewood Cliffs NJ:
Prentice Hall.

46

Heart and Circulation

Underwood, R. 1992. Magnetic resonance imaging of the cardiovascular system. In J.

Reiber and E. Ernst (eds.) Cardiovascular Nuclear Medicine and MRI. Dordrecht:
Kluwer Academic.
Webster, J. G. 1998. Measurement of flow and volume of blood. In J. G. Webster (ed.)
Medical Instrumentation: Application and Design. 3rd ed. New York: John Wiley &
Sons.
Zorn, E. A., Wilson, M. B., Angel, J. J., Zanella, J. and Alpert, B. S. 1997. Validation of
an automated arterial tonometry monitor using Association for the Advancement of
Medical Instrumentation standards. Blood Press. Monit. 2 (4): 1858.

8.13 Problems
8.1
8.2
8.3
8.4
8.5
8.6
8.7
8.8
8.9

8.10

Sketch the anatomy of the heart chambers and valves and show the direction of
blood flow.
Explain why the wall of the left ventricle is thicker and more muscular then the
wall of the right ventricle.
Describe the relation between the arterial pressure and the left ventricular pressure during a cardiac cycle.
Explain why the AV node delays the wave of excitation.
Draw a typical ECG waveform, label all waves, and explain what is happening
within the heart muscle during each wave.
Explain why different pairs of electrodes are used on specific locations on the
body for an ECG.
An ECG has a scalar magnitude of 0.8 mV on lead I and a scalar magnitude of
0.2 mV onlead III, calculate the scalar magnitude of lead II.
Given the stroke volume = 80 mL, LV end-diastolic volume = 51 mL , and the
heart rate = 80 bpm, calculate the LV ejection fraction and the cardiac output.
Use the Fick method to calculate the cardiac output (in mL/min) of a patient
whose arterial concentration of O2 is 0.20 L/L, venous concentration of O2 is
0.13 L/L, and the O2 consumption is 0.3 L/min.
Calculate the cardiac output using the thermodilution method from the following
data:

8.11
8.12
8.13
8.14

t1

| T
0

| dt = 3 s C , Q = 1250 J, b = 1060 kg/m3, and cb = 3640

J/(kgK).
Explain why and how we measure cardiac output by the thermal and dye dilution techniques and compare the two methods.
A Doppler-ultrasonic blood flowmeter has a carrier frequency of 7.5 MHz, with
a transducer angle of 30. The velocity of the sound is 1500 m/s and the audio
frequency is 12.6 kHz. Calculate the blood velocity.
Describe the cardiac sounds, their origins, and the time they occur in the cardiac
cycle.
An artery has a diameter of 3 mm. The viscosity of the blood is 0.0028 Pas at
37 C, and the mean arterial blood pressure is 90 mmHg at one end and 0 mmHg

8.12 References

8.15
8.16
8.17
8.18

8.19
8.20
8.21
8.22

47

at the other end. Calculate the blood flow rate in mL/min, assuming that the
length of the artery equals 0.5 m.
Sketch an electromagnetic flowmeter and explain its principle of operation.
Sketch an ultrasonic flowmeter and explain its principle of operation.
Draw the block diagram for and describe the automatic indirect auscultatory
peripheral blood pressure measurement system.
Draw the block diagram for the automatic indirect oscillometric (not auscultatory) peripheral blood pressure measurement system. Show all pneumatic connections between parts. Show location of all sensors and explain how they work.
Describe the measurement cycle. Sketch the resulting waveforms. Explain how
the signal processing identifies significant pressures.
Explain why and how we measure direct cardiac pressure using a catheter
(catheter-type and catheter tip).
Explain why and how we measure vessel pressure using arterial tonometry.
Explain why and how we measure vessel pressure distensibility.
Explain why and how we measure vessel volume flow.

9
Lung, Kidney, Bone and Skin
Shilpa Sawale

This chapter describes four major organs in the body: lung, kidney, bone, and skin. Each
of these organs has its own properties and functions: the lungs help us respire, the
kidneys help clean the blood, bone supports the body, and skin protects the body.
Measurements are performed to confirm whether each of these organs is functioning
properly, and also to measure some of their properties.

9.1 Lung
The exchange of gases in any biological system is called respiration. To sustain life, the
human body needs oxygen, which is utilized in cells with other essential nutrients during
the metabolic oxidation process. Carbon dioxide is a by-product of cellular metabolism.
The hemoglobin in the blood is the dominant transport mechanism by which oxygen is
brought to cells. Carbon dioxide produced by cells is dissolved in the blood plasma and
carried off for disposal via the lungs. The most important function of the lungs is to
supply tissue with adequate oxygen and to remove excess carbon dioxide.
To properly exchange gases during inspiration, alveolar pressure must be less
than the atmospheric pressure. There are two ways of producing the pressure difference
necessary for inspiratory flow: (1) the alveolar pressure can be reduced below
atmospheric. This is called natural or negative pressure breathing; (2) the atmospheric
pressure can be raised above normal and therefore above the resting alveolar pressure.
This is called positive pressure breathing.
Normal breathing is accomplished by active contraction of inspiratory muscles,
which enlarges the thorax. This further lowers intrathoracic pressure, which normally is
less than atmospheric pressure and hence, the air at atmospheric pressure flows through
the nose, mouth and trachea to the lungs.
The process of bulk gas transport into and out of the lungs and the diffusion of
gases across the alveolar membrane is known as pulmonary function. Tests performed to
determine parameters of system efficiency are called Pulmonary Function Tests (PFTs)
(Feinberg, 1986). PFTs do not tell us the ultimate answer. For example, finding a

Lung, Kidney, Bone and Skin

decreased lung volume reveals restriction, but not the cause of it. However, in several
circumstances PFTs are a useful and powerful tool: diagnosing a lung disease and finding
the extent of the abnormality, following a patient during the course of a disease to
determine the efficiency of the treatment or the need for supplemental oxygen and
mechanical ventilation, determining if preoperative patients can withstand the surgery,
assessing disability, and deciding whether an individual can perform an occupational task
requiring a certain work load (Petrini, 1988).

9.2 Pulmonary Volume

The most commonly used measures of the mechanical status of the ventilatory system are
the absolute volume and changes of volume of gas space in the lungs achieved during
various breathing maneuvers. Pulmonary volumes have been divided according to
physiological limits (Petrini, 1988). The common terms used to describe various
capacities (capacities refer to quantities that include more than one volume) are as
follows:
1.

Total Lung Capacity (TLC) is the largest volume to which the subjects lung can
be voluntarily expanded.
2. Residual Volume (RV) is the smallest volume to which the subject can slowly
deflate his or her lung.
3. Functional Residual Capacity (FRC) is the resting lung volume achieved at the
end of normal expiration. It is the point where inward pulling forces of the lung
and outward pulling forces of the chest wall are equal in magnitude.
4. Tidal Volume (TV) is the volume inspired and expired during normal breathing.
All of the lung capacities are based on these four parameters (Eqs. (9.1) to (9.4)).
IRV is the inspiratory reserve volume. ERV is the expiratory reserve volume.
Inspiratory capacity (IC) = IRV + TV

(9.1)

Functional residual capacity (FRC) = RV + ERV

(9.2)

Vital capacity (VC) = IC + ERV

(9.3)

Total lung capacity (TLC) = VC + RV

(9.4)

In dealing with volumes of gases, the conditions under which the values are
reported must be well defined and carefully controlled because gases undergo large
changes under different thermodynamic conditions. These volumes and specific
capacities, represented in Figure 9.1, have led to the development of specific tests to
quantify the status of the pulmonary system.

Lung, Kidney, Bone and Skin

TLC
IRV
IC
TV

VC
ERV
FRC
RV

Figure 9.1 The spirometer measures lung capacities and lung volumes. Because the subject cannot
make the lung volume equal to zero, the spirometer cannot measure RV and FRC.

9.2.1 Measurement of pulmonary volume

The measurement of changes in the lung volume have been approached in two different
ways. The first involves measurement of the gas through the airway opening. This is
referred to as spirometry. The second method is to measure changes in the volume of the
gas space within the body using plethysmographic techniques.
Spirometer
The spirometer was one of the early instruments used for estimating respiratory volumes.
It collects the gas passing through the airway opening and computes the volume it had
occupied within the lungs. The tests performed by the spirometer can be classified into
two major groups: single-breath tests and multiple-breath tests. There are three types of
tests under the category of single-breath tests: tests that measure expired volume only,
tests that measure expired volume in a unit time, and tests that measure expired
volume/time. Maximal Voluntary Ventilation (MVV) is measured in the category of
multiple-breath tests (Feinberg, 1986).
A spirometer is an expandable compartment consisting of a movable, statically
counterbalanced, rigid chamber or bell, a stationary base, and a dynamic seal between
them. This seal is often water. Changes in internal volume of the spirometer are
proportional to the displacement of the bell. This motion can be observed on a calibrated
scale or a rotating drum (kymograph) as shown in Figure 9.2.

9.2 Pulmonary Volume

Counterweight
Pulley

Bell
Kymograph

Water seal
One-way
Valves

Soda-lime
cannister
Mouthpiece

Tube to the
patient

Figure 9.2 In the water sealed spirometer, expired CO2 is removed in the soda-lime cannister.

The mouthpiece of the spirometer is placed in the mouth of the subject whose
nose is blocked. As the gas moves into and out of the spirometer, the pressure of the gas
in the spirometer changes, causing the bell to move.
The system can be modeled as two gas compartments connected such that the
number of moles of gas lost by the lungs through the airway opening is equal and
opposite to the number gained by the spirometer. For rebreathing experiments, most
spirometer systems have a chemical absorber (soda lime) to prevent build up of carbon
dioxide. With the exception of water vapor in a saturated mixture, all gases encountered
during routine respiratory experiments obey the ideal-gas law during changes of state:

P=

N
RT = RT
V

(9.5)

where P is the pressure of the ideal gas, R is the universal gas constant, T is the absolute
temperature, is the mole density, which for a well-mixed compartment equals the ratio
of moles of gas N in the compartment to the compartment volume V.

Lung, Kidney, Bone and Skin

Spirometers should be able to measure at least 7 liters of gas with 3% accuracy.

They should be linear, have stability (no drift), low inertia, low resistance, and no
hysteresis. During slow breathing maneuvers, spirometers yield the output shown in
Figure 9.1, from which impared vital capacity can indicate disease.
Over a few minutes, the O2 in the spirometer is consumed. The expired CO2 is
absorbed by the soda lime cannister. Thus the bell slowly decends. The average slope of
the spirometer output yields the O2 consumption vs. time, as required for basal
metabolism tests and the Fick method of cardiac output measurement.
The soda lime cannister adds pneumatic resistance, so is removed when
measuring maximal voluntary ventilation. The subject breathes as hard as possible for 15
s and the peak-to-peak sum of all breaths multiplied by four yields the volume breathed
per minute.
Forced expiratory volume (FEV) measures airway restriction. The subject
inhales as much as possible, then blows out hard. A normal subject can exhale 83% of
vital capacity in 1 s whereas a subject with restricted airways takes longer.
Body plethysmograph
Because alveolar pressure cannot be measured directly, an indirect method has been
developed using a device known as a total body plethysmograph. This device is a
volumetric displacement box into which the patient is placed. The box is closed and
sealed so that no gas can enter or escape except through the connection leading to a
patient mouthpiece or mask. When the outlet connection is attached to a spirometer,
pneumotachometer, or respiratory function computer, a wide variety of measurements
can be easily made. It is used clinically not only to evaluate the absolute volume and
volume changes of the lungs, but also to provide a continuous estimate of alveolar
pressure, from which airway resistance can be found. Figure 9.3 shows a total body
plethysmograph.

9.2 Pulmonary Volume

Body plethysmograph
PL

Flow Q

Lung

PB

PL
Pneumotachometer

VL
PB

Shutter

VB

Figure 9.3 The total body plethysmograph measures lung volume with the shutter closed and the
airway resistance via a pneumotachometer with the shutter open.

In order to compute the alveolar pressure using plethysmograph pressure

response to thoracic volume changes, the Volume of Thoracic Gas (VTG) must be known.
A patient sits inside the plethysmograph with a pneumotachometer flow meter in his or
her mouth. Both mouth and box pressures are displayed on the x and y axes of the
oscilloscope for one breath cycle against a total mechanical obstruction to airflow. This is
accomplished using a solenoid operated shutter to block the pneumotachometer. When a
patient breathes against a closed shutter, the air flow is zero, and thus the pressure
measured at the mouth is the same as the alveolar pressure (Feinberg, 1986).
The tissues of the body are composed of liquids and solids and can be
considered incompressible when compared with the gas in the lungs. During breathing
movements, the tissues change shape, not volume. Also because the volume of the
plethysmograph is constant, changes in the volume of the plethysmograph are zero.
Consequently the change in the volume of the gas in the lungs VL is equal and opposite
to the change in the volume of the gas space in the box VB.
During compression, VL = VB. There is a large change in pressure of the
lung PL, while there is a small change in pressure in the box PB. As a first order
approximation, VL/VB = PB/PL. The partial pressure of water vapor in the lung must
be subtracted from the total pressure to yield compressible dry gas. The lung is
isothermal whereas the box is adiabatic. All this can be calibrated by plotting PB/PL
during breathing with the shutter closed, from which we learn VL (Primiano, 1998). After
calibration, we can measure PB and use it to predict PL. Then we can open the shutter
to obtain the airway resistance
RAW = PL/Q

(9.6)

Lung, Kidney, Bone and Skin

where Q is the flow measured by the pneumotachometer.

Impedance plethysmography
Impedance plethysmography is the measurement of physiologic volume changes by
measuring electric impedance. We can calculate the fixed impedance from Eq. (2.8).
Each time the heart ejects blood into the arteries, their diameter increases about
7%. Because the arteries have a regular cylindrical shape, we can calculate their
longitudinal change in impedance. We shall follow Swansons derivation, to develop the
equations used in impedance plethysmography. The derivation is based on three
assumptions: the expansion of vessels is uniform (this may not be the case in diseased
vessels), the resistivity of the blood does not change, and lines of current are parallel to
the arteries (Webster, 1998).
The shunting impedance of the blood, Zb, is due to the additional blood volume
V that increases the cross-sectional area A.

Zb =

bL
A

(9.7)

L
Electrodes
A
Meter
Figure 9.4 A model for two electrode impedance plethysmography for cylindrical vessels.

V can be calculated using the following equation:

V =

b L2 Z
Z2

(9.8)

If the assumptions are valid, the above equation shows that we can calculate V
from b and other quantities that are easily measured.
However, the lung does not have a regular shape, so we cannot use equations to
predict volume changes. In respiratory impedance plethysmography, respiration is
electrically monitored by measuring changes in chest impedance. The electric impedance

9.2 Pulmonary Volume

of the thoracic cavity changes with breathing movements and can be sensed in order to
monitor ventilatory activity. Ventilation produces changes in the shape and content of the
chest that are electrically observable as small impedance variations. A 100 kHz
impedance measurement between electrodes on each side of the chest yields a waveform
that follows ventilation. Electrodes can be easily attached to a segment of the chest and
the resulting impedance of the chest measured. As resistivity increases in response to
increased air in the chest, the impedance of the chest increases. It is useful for monitoring
breathing of infants as in infant apnea monitors.
Inductance plethysmography
Respiratory inductive plethysmography continuously monitors the motions of the chest
wall that are associated with changes in thoracic volume. The variable inductance sensor
uses a single loop of wire attached in a zigzag pattern to its own compliant belt. It is
excited by a low-level radio-frequency signal. Changes in the loops cross-sectional area
produce corresponding changes in self-inductance. An output is obtained proportional to
the local cross-sectional area of the segment of the chest wall that is encircled by the loop,
after demodulation. Usually two separate instruments are used, one around the chest and
the other around the abdomen. Their sum is a more accurate estimate of lung volume that
that obtained from a single loop. It is used during sleep studies to diagnose sleep apnea
(Kryger, 1994).

9.3 Pulmonary Flow

When the lung changes volume during breathing, a mass of gas is transported through the
airway opening by convective flow. Measurement of the variables associated with the
movement of this gas is very important. The volume flow rate and the time integral of
volume flow rate are used to estimate rate of change of lung volume and changes of lung
volume. Convective flow occurs as a result of a difference in pressure between two points.
From the relationship between pressure difference and volume-flow rate through a
system, measurement of the difference in pressure yields an estimate of flow.
9.3.1 Measurement of pulmonary flow
Pneumotachometer
The most commonly used instrument for measurement of flow is the pneumotachometer.
This instrument is designed to measure respiratory gas flow during inspiration and
expiration. It is based on the pneumatic equivalent of Ohms law:

10

Lung, Kidney, Bone and Skin

Q(flow) =

P (difference in pressure)
R (fixed resistance of pneumotachometer)

(9.9)

The flow resistors have approximately linear pressureflow relationships. Flowresistance pneumotachometers are easy to use and can distinguish the direction of
alternating flows. They also have sufficient accuracy, sensitivity, linearity, and frequency
response for most clinical applications. Even though other flow resistance elements are
incorporated in pneumotachometers, the most common are either one or more fine mesh
screens placed perpendicular to flow or a tightly packed bundle of capillary tubes or
channels with its axis parallel to flow. These devices exhibit a linear pressure dropflow
relationship for a wide range of steady flows, with the pressure drop nearly in phase with
the flow.
This element is mounted in a conduit of circular cross section. The pressure drop
is measured across the resistance element at the wall of the conduit. This pressure
difference is measured by a differential pressure sensor and is used as a measure of airflow through it.
The prevention of water vapor condensation in the pneumotachometer is very
important because the capillary tubes and the screen pores are easily blocked by liquid
water, which decreases the effective cross-sectional area of the flow element and causes a
change in resistance. Also, as water condenses, the concentration of the gas mixture
changes. To avoid these problems, a common practice is to heat the pneumotachometer
element. This can be done using various techniques. Usually a pneumotachometer is
provided with an electric resistance heater. Also, the screen of the pneumotachometer can
be heated by passing a current through it, heated wires can be placed inside or heating
tape or other electric heat source can be wrapped around any conduit that carries expired
gas (Primiano, 1998).
P

Fine mesh
screen

Q
(a)

Closely packed
channels

Q
(b)

Figure 9.5 A pneumotachometer measures flow from a pressure drop P across resistance
elements such as (a) a fine mesh screen or (b) capillary tubes or channels.

9.4 Pulmonary Diffusion

9.3 Pulmonary Flow

11

The amount of gas transferred from the lung to blood is a function of the pressure
gradient and the diffusion capacity (DL) if the gas is diffusion limited. The equation is
Vg = DL(PAg Pcg)

(9.10)

where Vg is the rate of transfer of gas (g) from alveoli (A) to the capillaries (c), and P is
the pressure. DL is the diffusing capacity in mL/(minmmHg) and includes the
parameters of the surface area of gas transfer, the thickness of the membrane across
which the gas is transferred, and the properties of the gas such as the molecular weight
and solubility. Since DL is measured from the gas side, Vg is negative since the gas is
moving away from the lungs (Primiano, 1998).
Carbon dioxide diffuses across the alveolar membrane much more easily than
oxygen, so a diffusion defect affects oxygen transfer first. Because of this and because of
the important role oxygen plays in sustaining life, it is important to evaluate a diffusion
capacity for oxygen. But obtaining the PaO2 requires a sample of arterial blood. To avoid
this, clinicians use carbon monoxide as the tracer gas as its properties are quite close to
those of oxygen, and so its diffusion capacity provides a reasonable estimate of the
diffusion capacity of oxygen. In addition, it also has affinity for hemoglobin at low
concentrations; hence all carbon monoxide that enters the blood chemically combines
with the hemoglobin in the red blood cells. One of the many methods that have been used
to find out the diffusion capacity of carbon monoxide involves a single breath technique.
The subject inspires a mixture of air, 0.3% (or less) carbon monoxide and helium
(approximately 10%) from RV to TLC. The subject holds his breath at TLC for about
10 s and then forcefully exhales down to RV. Although it requires subject cooperation, it
can be performed quickly and can be repeated easily. Also, it does not require samples of
arterial blood flow. The computation of DLCO from measurements made during this
single breath technique are based on a one compartment model of the lung. If a wellmixed alveolar compartment is filled with a mixture of gases containing some initial
alveolar fraction FACO, then during breath holding with the airway open, the CO
diffuses into the blood in the pulmonary capillaries, and the alveolar FACO decreases
exponentially with time (Primiano, 1998).

D CO ( Patm PA H 2 O )(t 2 t1 )
FA CO(t 2 ) = FA CO(t1 ) exp L

VA

(9.11)
9.5 Pulmonary Airway Resistance
Resistance to air flow is determined by the same factors governing the flow of fluid of
low viscosity in tubes. Resistance to air flow is a calculated quantity instead of a directly
measured one. The equation used is as follows

12

Lung, Kidney, Bone and Skin

Pulmonary resistance =

Mouth pressure Intrathoracic pressure

Flow

(9.12)

A noninvasive method of measuring airway resistance is the body

plethysmograph. An invasive method for measuring intrathoracic or intraesophageal
pressure involves passing an air-filled catheter with a small latex balloon on its end
through the nose into the esophagus. The balloon should be located such that pressure
changes due to motions of other organs competing for space in the thoracic cavity are
minimized. A region below the upper third of the thoracic esophagus gives low cardiac
interference and provides pressure variations that correspond well in magnitude and
phase with directly measured representative changes in pleural pressure. The
correspondence decreases as lung volume approaches the minimum achievable volume
(RV). The frequency response of an esophageal balloon pressure-measurement system
depends on the mechanical properties and dimensions of the pressure sensor, the catheter,
the balloon, and the gas within the system. The use of helium instead of air can extend
the usable frequency range of these systems (Primiano, 1998).

9.6 Kidney
Kidneys are paired bean-shaped organs lying on either side of the spine in the upper part
of the abdomen. Each kidney is connected to the aorta and vena cava by a single renal
artery and vein. By means of these blood vessels, the kidneys filter about 1200 mL/min
of blood. Each kidney consists of approximately a million microscopic units called
nephrons, which are made up of two components: a glomerulus and a tubule. The key
separation functions of the kidney are: elimination of water-soluble nitrogenous endproducts of protein metabolism, maintainence of electrolyte balance in body fluids and
elimination of excess electrolytes, contribution to the obligatory water loss and discharge
excess water in the urine, and maintenance of the acidbase balance in body fluids and
tissues. Normal renal function is to remove water, electrolyte, and soluble waste products
from the blood stream. The kidneys also provide regulatory mechanisms for the control
of volume, osmolality, electrolyte and nonelectrolyte composition, and pH of the body
fluids and tissues. Figure 9.6 shows the general anatomy of the kidney.

9.5 Pulmonary Airway Resistance

13

Ureter

Kidneys

Bladder
Urethra

Figure 9.6 The kidneys excrete urine through the ureters to the bladder, where it is voided through
the urethra.

9.7 Kidney Clearance

9.7.1 Creatinine

The glomerulus acts as a filter and separates water and crystalloids from the blood plasma,
while retaining blood cells and proteins. Nonselective glomerular filtration is followed by
the selective re-absorption in the renal tubes. Of the 180 L of water filtered each day,
perhaps 1 to 2 L are excreted. Likewise, all glucose, amino acids, and the small amount
of protein filtered together with most of the sodium, chloride, bicarbonate and calcium is
re-absorbed. The re-absorption of these ions is precisely regulated to return to the body
the exact quantity required to maintain the composition of the extracellular fluid.
Creatinine, a byproduct of phosphate metabolism in muscle, is frequently used as a
measure of GFR (Glomerular Filtration Rate) because it is freely filtered and not
reabsorbed: the quantity crossing the glomerular filter is thus same as the steady state
excretion of urine. For creatinine, the clearance (or volume of plasma that contains the
quantity excreted in unit time) is the same as GFR and is expressed in L/day (Gregory,
1988).

14

Lung, Kidney, Bone and Skin

GFR C =

UV
P

(9.13)

where U is the urinary concentration, P is the plasma concentration, V is the urine flow
rate and C is the creatine clearance. The normal glomerular filtration rate averages 120
mL/min, about 116 mL/min is re-absorbed, and about 4 mL/min is excreted as urine.

9.8 Kidney Imaging

9.8.1 Pyelogram

The main components of the urinary tract have either a tubular or cavitational structure
and their delineation with radio-opaque medium is readily possible. In this method the
radioactive dye is injected into the urethral passage. After a fixed amount of time has
passed, the X ray is taken. A satisfactory pyelogram demonstrates the kidneys position,
size, shape, its cup-shaped cavity, and its functional ability to concentrate the opaque
medium.
The one shot Intravenous Pyelogram (IVP) is performed when there is suspected
or potential injury to the kidney or the ureter. If one kidney is injured it is important to
know if the other kidney is functioning normally. The one shot intravenous pyelogram is
a very good way to determine this. The presence of foreign bodies, retroperitonial
hematomas, and spine fractures can also be determined by the one shot IVP. In this
technique 1 mL/kg of nonionic radiocontrast agent is usually injected as a rapid
intravenous bolus. Five minutes after the injection of contrast agent, an anteroposterior
radiograph of the abdomen is obtained. In a hypotensive patient, the one shot IVP may
not be helpful as there may be no secretion of the contrast due to decreased renal blood
flow.

9.9 Hemodialysis
Dialysis is necessary when the GFR falls below 5 to 10% of normal. There are two types
of renal failure: acute (days or weeks) and chronic (months or years). Chronic renal
failure is a group of conditions in which renal function is slowly eroded and from which
recovery is very unusual, while in acute renal failure recovery is within a few days or
weeks. Thus dialysis is a temporary measure in acute renal failure but in chronic renal
failure it is needed for life or until a kidney is successfully transplanted. Hemodialysis is
a process in which blood is continuously circulated in contact with a permeable
membrane, while a large volume of balanced electrolyte solution circulates on the other
side of the membrane. Diffusion of dissolved substances from one stream to the other
removes molecules that are in excess in the blood and replaces those for which there is a
deficiency. Three processes occur simultaneously during dialysisosmosis, diffusion
and ultrafiltration. Osmosis is movement of fluid across membranes from a lower

9.9 Hemodialysis

15

concentration of solutes to a higher concentration of solutes. Diffusion is the passage of

particles through a semipermeable membrane from higher concentration to a lower
concentration. Ultrafiltration is the bulk flow of solution brought about by hydrostatic
pressure across the semipermeable membrane.
The amount of fluid removed from the blood each hour is calculated using the
formula
mL/h = TMP KUF

(9.14)

where TMP is the Transmembrane Pressure (mmHg) and KUF is the Coefficient of
Ultrafiltration of the dialyser ((mL/h)/mmHg). Transmembrane pressure is the arithmetic
difference between the pressure on the blood side of the dialyser (usually positive) and
the dialysate pressure (usually negative). The negative pressure is calculated using the
formula

NP = TMP

AP + VP
2

(9.15)

where NP is the negative pressure, AP is the arterial pressure, and VP is the venous
pressure. The transmembrane pressure can be calculated using the formula:
TMP =

Total fluid to be removed hours on dialysis

KUF of a dialyser

(9.16)

To perform dialysis, an artificial kidney is used, which is commonly known as a

hemodialyser, dialyser, or filter. This dialyser can be a coil type, parallel plate type, or
hollow fiber type. In the hollow fiber type of dialyser, which is the most commonly used
dialyser, the blood flows through the hollow fibers and the dialysate streams past the
outside of the fibers. The blood and the dialysate flow counter current to maximize
efficiency of solute transfer. Figure 9.7 shows a typical dialyser.

16

Lung, Kidney, Bone and Skin

Blood from the patient

Dialysate output

Dialyser
Hollow fibers
Dialysate input

Adjustable constriction
Hollow fibers
Cleansed blood to patient

(a)

(b)

Figure 9.7 Typical dialyser (a) indicating the counter current flow of the blood and the dialysate (b)
Cross-sectional view of the dialyser.

Hemodialysis membranes vary in chemical composition, transport properties,

and biocompatibity. These membranes are fabricated from three classes of biomaterials:
regenerated cellulose, modified cellulose, and synthetics. Chapter 4 describes some
details of biomaterials. Cuprophan (cupramonium cellulose), cellulose acetate,
polyacrylonitrile, polymethyl methacrylate polycarbonate, and polysulphate are some
other commonly used materials to prepare dialysis membranes.
Figure 9.8 shows a schematic of a hemodialysis machine (Gregory, 1988). The
blood from the patient is passed into the dialyser with the help of a roller blood pump.
Blood clots rapidly when it comes in contact with most foreign substances. To avoid
blood clots an anticoagulation agent is added. Heparin is the most commonly used
anticoagulation agent. The blood passes through the dialyser where the diffusion takes
place between the dialysate and the blood. Clean blood that comes out of the dialyser is
then injected into the patient. The blood is taken out from the arterial side of the patient
and is injected into the venous side. A venous bubble trap is inserted in the circuit to
make sure that the blood going back to the patient does not have an air bubble in it. An
ultrasonic sensor or photocell is used for bubble detection. The ultrasonic sensor monitors
any change in sound transmission and the photocell monitors any change in light
transmission. Whenever there is air on the venous side, the signals being transmitted are
interrupted and trigger an alarm. The pressure on the venous side is also monitored using
a pressure gage.

9.9 Hemodialysis

17

Heparin infusion
Blood from
patient

Dialyser

Venous pressure gage

Pump
Air/foam detector
Blood returning to
patient

Blood leak detector

Dialysate
pressure gage

Dialysate flow
meter

Pure water

Dialysate
pump

Spent dialysate
to waste

Adjustable
constriction

Heater

Conductivity and
pH cell

Concentrate

Proportioning
pump

Figure 9.8 Blood from an artery flows past a membrane in the hemodialysis machine, then returns
to a vein.

Proportioning pumps first mix the dialysate concentrate with water until a
desired concentration is reached. The dialysate temperature, conductivity, pH, and flow
are monitored and controlled. The dialysate is warmed by a heater to near normal body
temperature. The high temperatures may cause hemolysis (rupture of RBCs). The
negative pressure of the dialysate is usually generated by having the dialysate pump draw
the dialysate through the dialyser and by placing an adjustable constriction upstream of
the dialyser as shown in Figure 9.7. The pH of the dialysate must match the acid/base
balance of the patients blood. The dialysate is then passed through the dialyser in the
opposite direction to blood flow. The dialysate coming out of the dialyser is then passed
through a blood leak detector to check if there has been a rupture of the membrane and if
any blood has leaked into the dialysate circuit side.
The blood leak detector uses a beam of light that shines through the used
dialysate and into a photocell. Normally the dialysate is clear and lets light pass through.
Even a tiny amount of blood causes a break between the light beam and the photocell.
Any break in the transmission of the light beam triggers an alarm. The spent dialysate is
then drained off.

9.10 Peritoneal Dialysis

18

Lung, Kidney, Bone and Skin

Each organ in the abdominal or peritoneal cavity is contained within a visceral

peritoneum membrane. The cavity is lined with a parietal peritoneum membrane. The
linings are thin layers of flattened mesothelial cells. The average adult peritoneal cavity
has a surface area of 1 to 2 m2. Peritoneal dialysis takes advantage of the enclosed
volume. During the process the peritoneal cavity is filled with dialysate through a
surgically implanted catheter. Through diffusion, urea, potassium, excess electrolytes and
other wastes cross the peritoneal membrane into the dialysate. The diffusion continues
until an equilibrium concentration is reached between the plasma and the dialysate. The
dialysate is periodically drained and replaced with fresh dialysate to continue the process.
In addition to waste removal, the kidneys maintain the bodys water level. For excess
water to diffuse across the membrane the dialysate must be made hypertonic by adding a
high concentration of glucose to it. Glucose can readily cross the membrane and be
absorbed by the body while water filters across the membrane into the dialysate. The
patient may absorb 200 to 500 kcal during peritoneal dialysis.
Peritoneal dialysis provides the patient with added independence. The patient
controls the process. Unlike hemodialysis, they are not required to spend 4 to 6 h three
times per week in a clinic. However, during the process the patient must be connected to
a control system to weigh, warm, and refresh the dialysate. The process is typically
performed three times a week and may take 8 to 10 h. To improve patient independence,
manufacturers of systems aim to provide a low-cost, portable system. Many patients
prefer an automated system that performs the dialysis while they sleep. Figure 9.9 shows
a peritoneal dialysis system.
Heating element
Valve

Pump
Check valve
Pump

Dialysate
supply

Catheter

Peritoneal
cavity

Body
fluids
Spent
dialysate

Figure 9.9 A simple schematic of a peritoneal dialysis system.

9.10 Peritoneal Dialysis

19

To provide a mobile system, the control hardware, dialysate supply and spent
dialysate are commonly mounted on a wheeled stand. The dialysate supply is normally
positioned high on the stand, above the abdomen, to provide a gravity flow. The spent
dialysate collection is positioned at the bottom of the stand for same reason. Most
systems are equipped with peristaltic pumps to control flow. Traveling or portable
systems may exclusively use gravity to provide a smaller, lighter system. The dialysate is
warmed by the control electronics prior to flowing through the catheter and into the
peritoneal cavity to prevent thermal shock. After approximately 30 min, the dialysate is
pumped out of the cavity and into a collection bag. The weight of the dialysate supply
and spent dialysate is monitored to determine the amount of fluid and waste removed
from the body.
The key measurement in the peritoneal dialysis process is the weight of the
fluids. It is not practical to attempt to measure the change in concentration of the
dialysate. Therefore, the amount of fluid pumped into the peritoneal cavity and the
amount of fluid removed are measured as the means of monitoring the amount of water
and waste diffused from the body fluids. The circuit in Figure 9.10 shows a possible
means of measuring the fluid weight.
15 V

13 k

Rf
Ri

2 k

vi

vo

Figure 9.10 Dialysate weight measuring circuit.

The weight of the fluid determines the position of the wiper on a potentiometer.
The fluid may be suspended from a spring-loaded mechanism that varies the wiper
position. In this case, the value of the variable 2 k resistor is dependent upon the weight
of the fluid and has been installed in a voltage divider powered by the 15 V power supply
to the op amp. A change in the resistance varies the input voltage, vi, to the non-inverting
op amp circuit. The input voltage is amplified by (Rf + Ri)/Ri (see section 2.2). The
amplification is limited by the 12 V saturation of the op amp. Acceptable resistance
values for Ri and Rf are 1 k and 5 k, respectively. The maximum value of vi is 2 V,
which results in a 12 V output voltage. The output voltage can be sent to an analog-todigital converter and then to a CPU for data storage and processing.
The greater the osmotic pressure difference across the membrane, the more
diffusion occurs. As the pressure decreases, less diffusion occurs. The concentration of
body fluid waste is exponentially removed. It is important to understand that it is this
relationship that is used to determine the frequency with which the dialysate should be

20

Lung, Kidney, Bone and Skin

refreshed. If peritoneal dialysis can accurately be simplified into the model shown in
Figure 9.9 (two volumes of fluid separated by a membrane), a relationship can be
developed to determine the rate at which the body fluid wastes or solutes diffuse across
the membrane.
The concentrations of the solutes are defined as
C=

N
V

(9.17)

where C is the concentration of solute (molecules/m3), N is the number of particles in the

solute and V is the volume of the solute.
The solute flow rate through a membrane, Js, due to diffusion is defined as
J s = RT (C b C d )
N
N
J s = RT b d
V b Vd

(9.18)

where R is the gas constant (J/(molK)), T is the absolute temperature, 310 K, and is the
solute permeability of the peritoneal membrane (1 105 moles/(Ns)). The designations
of b and d stand for the body fluid and dialysate, respectively.
Through some algebra and knowing Js, the surface area of the peritoneum and
the initial concentrations, the concentration in the body determined as a function of time
is

Vd
Vb
C (t ) = C o
e t +

V b + Vd
V b + V d

(9.19)

This relationship can be plotted to determine at what point the dialysate should be
refreshed. Normally, 30 min cycles are done. After 30 min, the body fluid concentration
is approximately 0.9524Co using the values given above. The concentration after n cycles
is simply computed as
C (t ) = (0.9524) n Co
(9.20)

and may be used to estimate the reduction in body fluid solute concentration following
the dialysis. As mentioned previously, the rate of dialysis is monitored by weighing the
dialysate and spent dialysate.

9.11 Kidney Function

9.11 Kidney Function

21

9.11.1 Extracellular water by impedance

Human body composition is sometimes measured in order to assess nutritional

deficiencies, loss of water during dialysis, after severe burns, or any condition where the
fluid balance is in question. If the kidneys are not functioning properly, the body retains
water. The 24 h urine test is the most common diagnostic tool for renal failure. This test
measures the amount of urine a person produces in one day. Doctors may use it with
other tests to measure the excretion of certain substances, such as creatinine, into urine
over a 24 h period.
One other method to detect excess retention is bioelectrical impedance analysis
(BIA), which estimates total body water by passing about 800 A at 50 kHz between
electrodes on the wrists and ankles. A frequent equation used is
TBW = aHt2/R + bWt + c

(9.21)

where TBW is the total body water in m3, Ht is the height of the person in m, R is the
resistance in , and Wt is the weight in N. a, b, and c are constants determined by
calibrating against isotopic dilution techniques (see section 10.4.1) or using other
equations estimating lean body mass, from which body fat can be estimated.
This method is controversial. Patterson (1989) points out that wrist-to-ankle
measurement is influenced mostly by the impedance of the arm and leg and less than 5%
of the total impedance is contributed by the trunk, which contains half the body mass.
The National Institutes of Health (1994) concludes that there are many conditions in
critical illness for which conventional BIA is a poor measure of TBW.
Another method that is used to estimate extracellular water is by deduction.
There are three basic components of total body mass: water, lean body mass (muscle) and
body fat. Section 10.4 gives many methods of estimating body fat. Other methods
provide us with equations that estimate lean body mass. If one of each of these methods
are used to obtain an approximation for lean body mass and body fat, these can be
subtracted from the total body mass to obtain an estimate of the extracellular water.

9.12 Bones and Joints

Bones can last for centuries and in some cases for millions of years. They provide the
anthropologist with a means of tracing both the cultural and the physical development of
man. Because of the importance of bone to proper functioning of the body, a number of
medical specialists such as the dentist, orthopedic surgeon and radiologist are concerned
with the health of bone. Bones are also of interest to medical physicists and engineers
because they deal with engineering problems concerning static and dynamic loading
forces that occur during standing, walking, running, lifting and so forth. Nature has
solved these problems extremely well by varying the shapes of the various bones in the
skeleton and the type of bone tissue of which they are made. Bone has six important
functions in the body: support, locomotion, protection of various organs, storage of
chemicals, nourishment, and sound transmission (e.g. the middle ear). Bone consists of

22

Lung, Kidney, Bone and Skin

two different types of materials plus water. They are collagen, the major organic fraction,
which is about 40% of the weight of solid bone and 60% of its volume and bone mineral,
the so called inorganic component of the bone, which is about 60% of the weight of the
bone and 40% of its volume. Either of the components may be removed from the bone,
and in each case the remainder, composed of only collagen or bone mineral, looks like
the original bone. The collagen remainder is quite flexible, like a chunk of rubber, and
can even be bent into a loop. When collagen is removed from the bone, the bone mineral
remainder is very fragile and can be crushed with the fingers. Collagen is produced by the
osteoblastic cells and mineral is then formed on the collagen to produce bone (Cameron,
1978).
9.12.1 Bone mineral density

The strength of the bone depends to a large extent on the mass of bone mineral present. In
diseases like osteoporosis the bone mineral mass is considerably reduced. Up to a few
years ago osteoporosis was difficult to detect until a patient appeared with a broken hip or
a crushed vertebra. By that time it was too late to use preventive therapy. Thus bone
mineral is very important and commonly measured to detect bone diseases such as
osteoporosis. The bone mineral content of specimens can be measured by ashing the
specimen in a furnace or demineralizing it in a decalcifying solution (Ashman, 1989).
The most commonly used technique for noninvasively measuring bone mineral content in
the bone is dichromatic or dual photon absorptiometry (DPA).
In the early part of the 20th century, X rays were used to measure the amount of
bone mineral present in the bone. But there are some major problems with using an
ordinary X ray: the usual X-ray beam has many different bands of energy, and the
absorption of X rays by calcium varies rapidly with energy in this range of energies; the
relatively large beam contains much scattered radiation when it reaches the film; the film
is a poor tool for making quantitative measurements since it is nonlinear with respect to
both the amount and the energy of X rays. The net result of these problems is that a large
change in the bone mineral mass (30 to 50%) must occur between the taking of two X
rays of the same patient before a radiologist can be sure that there has been a change.
Figure 9.10 shows that in dual photon absorptiometry, three problems with the X-ray
technique are largely eliminated by using an X-ray source filtered to yield two
monoenergetic X-ray beams at about 30 keV and 70 keV, a narrow beam to minimize
scatter, and a scintillation detector that detects all photons and permits them to be sorted
by energy and counted individually. Tests are frequently made in the spine, hip, and
forearm but can be done on the entire body.

9.12 Bones and Joints

23

Scintillation
detector

Dual beam
Soft tissue
Bone

X-ray dual beam

radiation source
Figure 9.10 In a dual photon absorptiometer, an X-ray source is filtered to emit at two discrete
energies.

Hologic (2000) improves clinical bone densitometry by integrating bone mineral

density (BMD) with Instant Vertebral Assessment (IVA). IVA, possible only with fanbeam technology, generates a high-resolution image of the entire lateral spine in just 10 s,
enabling physicians to visually assess vertebral status for a more accurate determination
of fracture risk than just BMD alone. A different instrument measures the speed of sound
(SOS, in m/s) and broadband ultrasonic attenuation (BUA, in dB/MHz) of an ultrasound
beam passed through the heel, and combines these results to obtain the Quantitative
Ultrasound Index (QUI). The output is also expressed as a T-score and as an estimate of
the Bone Mineral Density (BMD, in g/cm2) of the heel.
9.12.2 Stress and strain
Tensile loads

When forces are applied to any solid object, the object is deformed from its original
dimensions. At the same time, internal forces are produced within the object. The relative
deformations created at any point are referred to as strains at that point. The internal force
intensities (force/area) are referred to as stresses at that point. When bone is subjected to
forces, these stresses and strains are introduced throughout the structure and can vary in a
very complex manner (Ashman, 1991).

24

Lung, Kidney, Bone and Skin

Figure 9.11(a) shows a cylindrical bar of length L and a constant cross-sectional

area A subject to a pure tensile force F. As the load is applied, the cylinder begins to
stretch. This situation can be described by an equation that describes the stretching of a
spring (Hookes law):
F = kx

(9.22)

where F is the applied force, x is the change in length or elongation of the spring, and k is
the spring constant or stiffness of the spring.
Force

c = Ultimate
tensile strength

Fult = Failure
load

F
L

Fy = Yield

Force, F
Stress,

y = Yield load strength

AE/L = Stiffness
E = Elastic
modulus
A, E
Deflection, L
Strain,

(a)

(b)

Figure 9.11 Tensile stress on a cylindrical bar causes tensile strain .

The analogous relation for stretching of the cylinder is

L =

FL
AE

(9.23)

where L is the elongation of the cylinder, L is the original unstretched length, A is the
cross-sectional area, F is the force and E is the elastic (Youngs) modulus (18 GPa for
compact bone).
Tensile, or uniaxial, strain, , can be calculated using the formula

L
L

(9.24)

Similarly, tensile stress, , is calculated using the formula

F
A

(9.25)

9.12 Bones and Joints

25

These tests can be performed on specimens with different lengths, cross-sectional areas
and under forces of varying magnitudes (Black, 1988).
Figure 9.11(b) shows that a cylinder of bone tested in tension yields a linear
region (also known as the elastic region) where the atoms of the bone are displaced only
slightly by reversible stretching of the interatomic bonds. This is followed by a nonlinear
region where yielding and internal damage occurs, often involving irreversible
rearrangement of the structure. After yielding, nonelastic deformation occurs until finally
failure or fracture results. The load at which yielding occurs is referred to as the yield
load, Fy. The load at which failure occurs is called the ultimate or failure load, Fult. This
curve describes the behavior of the structure since the curve differs for a different crosssectional area or different length. It also represents the mechanical behavior of the
material as opposed to the behavior of the structure. In the stressstrain curve, the
material yields at a stress level known as the yield strength and fractures at a stress level
known as the fracture strength or the ultimate tensile strength. Toughness is the ability to
absorb energy before failure and is calculated from the total area under the stressstrain
curve, expressed as energy per unit volume in J/m3 (Black, 1988).
The above quantities can be measured using any of the displacement type
sensors (e.g. strain gage, LVDT (Linear Variable Differential Transformer), capacitive,
and piezoelectric sensors). The most commonly used sensors are the strain gage and the
LVDT.
Shear Loads

When forces are applied parallel to a surface or along the edge of an object, the object
deforms in a way shown in Figure 9.12. The sides of the object perpendicular to the
forces stretch and shear stresses and strains result.

F
W

A = LW
F
Figure 9.12 Shear stress , causes shear strain .

Shear strain, , can be calculated using the formula

26

Lung, Kidney, Bone and Skin

L
L

(9.26)

Shear stress, , can be calculated using the formula

F
= F/A
A

(9.27)

where L is the distance of shear deformation, L is the original length, A is the crosssectional area and F is the acting force.
The shear modulus, G, can be calculated using and using the relationship:
G=

(9.28)

See more information about measuring shear stress and strain in Chapter 6.
9.12.3 Strain gage

The strain gage is a variable resistance sensor whose electric resistance is

R=

l
A

(9.29)

where R = resistance in , = resistivity in m, l = length of the wire in m, and A =

cross-sectional area in m2. An increase in length causes an increase in resistance. The
sensitivity is expressed by the gage factor
G=

R / R
/
= (1 + 2 ) +
L / L
L / L

(9.30)

where is Poissons ratio, which can be expressed as

D D
L L

(9.31)

where D is the diameter of the cylindrical specimen. Poissons ratio is the ratio between
the lateral strain and axial strain. When a uniaxial tensile load stretches a structure, it
increases the length and decreases the diameter.
Figure 9.13 shows four strain gage resistances that are connected to form a
Wheatstone bridge. As long as the strain remains well below the elastic limit of the strain
gage resistance, there is a wide range within which the increase in resistance is linearly
proportional to the increase in length.

9.12 Bones and Joints

R2

R1

R3

R4

27

vi
Rx

vo
Figure 9.13 Four strain gage resistances R1, R2, R3, and R4 are connected as a Wheatstone bridge.
vi is the applied voltage with the bottom terminal grounded. vo is the output voltage, which must
remain ungrounded and feeds a differential amplifier. Potentiometer Rx balances the bridge. See
section 2.1.6 for more information on Wheatstone bridges and potentiometers.

The types of strain gages are dictated by their construction. Unbonded strain
gages may be formed from fine wires that are stretched when strained, but the resulting
sensor is delicate. The wires resistance changes because of changes in the diameter,
length and resistivity. Bonded strain gages may be formed from wires with a plastic
backing, which are glued onto a structural element, such as carefully dried bone.
Integrated strain gages may be formed from impurities diffused into a silicon diaphragm,
which forms a rugged pressure sensor (Peura and Webster, 1998). A strain gage on a
metal spring is useful for measuring force within a uniaxial tensile test machine.
9.12.4 LVDT

A transformer is a device used to transfer electric energy from one circuit to another. It
usually consists of a pair of multiply wound, inductively coupled wire coils (inductors)
that facilitate the transfer with a change in voltage, current or phase. A linear variable
differential transformer (LVDT) is composed of a primary coil and two secondary coils
connected in series opposition, as shown in Figure 9.14. The ac excitation is typically 5 V
at 3 kHz. The coupling between these two coils is changed by the motion of the high
permeability magnetic alloy between them. When the alloy is symmetrically placed, the
two secondary voltages are equal and the output signal is zero. When the alloy moves up,
a greater voltage is transformed to the top secondary coil and the output voltage is
linearly proportional to the displacement. The LVDT is useful in determining the strain
on tendons and ligaments (Woo and Young, 1991).

28

Lung, Kidney, Bone and Skin

vo

x
vi

Secondary coil

Primary
coil

Secondary
coil
High permeability
alloy core
Figure 9.14 In a linear variable differential transformer, displacement of the high permeability
magnetic alloy changes the output voltage.

The most commonly used technique for measurement of stress and strain on
bone specimens is the Uniaxial tension test using the LVDT. Figure 9.15 shows the set up
for this test. It consists of one fixed and one moving head with attachments to grip the test
specimen. A specimen is placed and firmly fixed in the equipment, a tensile force of
known magnitude is applied through the moving head, and the corresponding elongation
is measured. Then using Eqs. (9.24) and (9.25), the uniaxial stress and strain can be
calculated.

9.12 Bones and Joints

29

Moving head
Grip
Specimen
LVDT

Bonded strain
gage

Grip
Fixed head

Figure 9.15 The uniaxial tension test measures force versus elongation.

9.12.5 Soft tissue strain

A direct method of measuring ligament strain is by mounting a strain gage load cell
within cadaveric knees (Markolf et al., 1990). A noncontact method for measuring
ligament strain is the video dimension analyzer (VDA) (Woo et al., 1990). Reference
lines are drawn on the specimen using stain and the test videotaped. The VDA system
tracks the reference lines and yields strain versus time. Tendon and ligament tissues
contain low-modulus elastin, which bears the majority of the load for strains up to about
0.07. At higher strains, collagen fibers, which possess a zigzag crimp, take an increasing
portion of the load, resulting in an upward curvature of the stressstrain curve (Black,
1988).
Proteoglycans are important components of the extracellular matrix of articular
cartilage and other soft tissues. Viscosity of proteoglycans extracted from cartilage can be
measured using a cone-on-plate viscometer described in section 6.7.1 (Zhu and Mow,
1990). The angle of the cone is 0.04 and the diameter of the plate 2a is 50 mm. The
shear rate = /, where is the rotational speed of the plate (rad/s). Apparent viscosity
app = 3T/(2a3), where T = torque. Another viscometer is the Ostwald capillary
viscometer, which calculates the coefficient of viscosity from the pressure gradient
dp/dL, the volume rate of flow Q, and the tube radius R using the equation =
[R4/(8Q)](dp/dL) (Fung, 1981). Another viscometer is the Couette coaxial viscometer in
which an outer rotating cylinder transmits torque through the test fluid to an inner coaxial
cylinder.

30

Lung, Kidney, Bone and Skin

9.12.6 Joint friction

Diarthrodial (synovial) joints have a large motion between the opposing bones. To
diagnose disease and to design artificial joints, we desire to measure friction between
them. The resistance to motion between two bodies in contact is given by frictional force
F = W, where is the coefficient of friction and W is the applied load (Black, 1988).
Surface friction comes either from adhesion of one surface to another due to roughness
on the two surfaces or from the viscosity of the sheared lubricant film between the two
surfaces. Lubrication of bone joints is an important factor in determining coefficient of
friction. Rheumatoid arthritis results in overproduction of synovial fluid in the joint and
commonly causes swollen joints. The synovial fluid is the lubricating fluid that is used by
the joints. The lubricating properties of the fluid depend on its viscosity; thin oil is less
viscous and a better lubricant than thick oil. The viscosity of synovial fluid decreases
under the large shear stresses found in the joint.
The coefficient of friction is measured in the laboratory using arthrotripsometers
(pendulum devices). Here a normal hip joint from a fresh cadaver is mounted upside
down with heavy weights pressing the head of femur into the socket. The weight on the
joint is varied to study the effect of different loads. The whole unit acts like a pendulum
with the joint serving as the pivot. From the rate of decrease of the amplitude with time,
the coefficient of friction is calculated. It can be concluded that fat in the cartilage helps
to reduce the coefficient of friction. When synovial fluid is removed, the coefficient of
friction is increased considerably. Figure 9.16 shows an arrangement for measuring the
coefficient of friction (Mow and Soslowsky, 1991).

9.12 Bones and Joints

31

Bone joint
(pivot)

Weights
Rocking movement
Figure 9.16 Decay of oscillation amplitude in the pendulum device permits calculation of the
coefficient of friction of a joint.

We wish to measure wear in joints of artificial materials such as ultra-highmolecular-weight polyethylene (UHMWPE) (Dowson, 1990). For polymeric materials,
the volume V = kPX produced by wear during sliding against metallic or ceramic
countersurfaces is proportional to the applied load P and the total sliding distance X. k is
a wear factor indicative of the wear resistance of a material.
9.12.7 Bone position

Bone position is important for calculating the loading forces that act on it. The most
complicated bones in the body on which these forces are calculated are in the spinal cord.
Calculating forces for other bones, such as the femur, is relatively easy. These loading
forces are static and dynamic in nature.
A goniometer is an electric potentiometer that can be attached to a joint to
measure its angle of rotation. For working details of the goniometer refer to Chapter 10.
Human joint and gross body motions can be measured by simple protractor type
goniometers, electrogoniometers, exoskeletal linkage devices, interrupted light or normal
and high speed photography, television-computer, X-ray or cineradiographic techniques,
sonic digitizers, photo-optical technique, and accelerometers (Engin, 1990).

32

Lung, Kidney, Bone and Skin

9.12.8 Bone strain-related potentials

Bending a slab of cortical or whole living or dead bone yields piezoelectric potentials of
about 10 mV, which we can measure using electrodes (Black, 1988). If the strain is
maintained, the potentials rapidly decay to a very low value, called the offset potential.
Some workers hypothesize that these potentials have a role in directing growth, healing,
and remodeling.

9.13 Skin
The primary function of the skin is to act as a buffer between the body and the outside
world. It absorbs sunlight, softens blows, insulates from heat and cold, retains fluid
within the body and keeps foreign organisms and extraneous chemicals out of it. It plays
an important role in bodys defense against infection. It also acts as the bodys radiator
and as such is the most significant element in the thermoregulation system.
9.13.1 Water loss

Transepidermal Water Loss (TWL or TEWL) refers to the rate at which water migrates
from the viable dermal tissues through the layers of the stratum corneum to the external
environment. In the absence of profuse sweating, the TWL is predominantly controlled
by the diffusion of water vapor in the stratum corneum caused by the difference in vapor
concentration between the inside and outside surfaces.
TWL measurements have been used in studying the restoration of barrier
function in wounded skin, as well as in evaluating occlusive properties of topical
preparations. TWL can be measured using the direct measurement technique or the
indirect measurement technique. Flow hygrometry is a commonly used technique for
direct measurement. The indirect measurement technique relies on establishing a
boundary air layer over skin of known geometry. The two commonly used techniques of
indirect measurement are the closed cup method and the open cup method (Marks, 1981).
Flow hygrometry

In flow hygrometry, the flux of water vapor out of a fixed area of skin is determined by
measuring the increase in water vapor concentration in the flowing gas stream. The
increase in humidity is read at a sensor output once steady-state conditions have been
reached. TWL can be calculated using the formula:
TWL =

K V R
A

(9.34)

9.13 Skin

33

where K is the instrument constant, V is the increase in the sensor output, R is the gas
flow rate and A the skin area isolated by the measuring chamber. TWL is usually
measured in mg/(cm2h). There are numerous possible sources of error in the flow
hygrometry system, which include uncertainty in the gas flow rate, uncertainty in the
actual area of skin exposed to flowing gas, absorption of water vapor in tubing
connecting the skin chamber with the sensor, and leaks in the seal of the chamber to the
skin site. These errors can be minimized by careful design of the system.
In this method, transfer of water vapor to the sensor is by convection, which
normally requires gas tanks or pumps, valves, and tubing in addition to the sensor and the
skin chamber. In contrast, indirect measurement techniques, which are described later,
require little auxiliary equipment. Figure 9.17 shows a flow hygrometer (Smallwood and
Thomas, 1985).

Meter

Gas
source

Humidity sensor

Flow rate = R

Skin
Figure 9.17 The flow hygrometer measures the increase in humidity of gas flowing over the skin.

Closed cup method

In the closed cup method, the humidity sensor is sealed into one end of a cylinder of
known length. The cylinder is placed on the skin, trapping a volume of unstirred air
between the skin and the detector as shown in Figure 9.18(a). Within a few seconds, the
sensor voltage begins to rise steadily followed by a gradual decrease in rate of change.
The application of diffusion principles to the water vapor in the volume of trapped air
predicts this behavior and shows that the TWL is directly proportional to the slope of the
transient linear portion of the detector output curve shown in Figure 9.18(b). The TWL is
found from:
TWL = K l

dv
dt

(9.35)

where K is an instrument calibration constant, l is the distance between the detector and
the skin surface, v is the detector voltage and t is the time. The closed chamber method

34

Lung, Kidney, Bone and Skin

does not permit recordings of continuous TWL because when the air inside the chamber
is saturated, skin evaporation ceases.
Sensor

Transient linear
portion

Time

Skin

(a)

(b)

Figure 9.18 The closed cup hygrometer: (a) configuration of measuring cup, (b) typical sensor
output curve.

Open cup method

Figure 9.19(a) shows the open cup hygrometer. The sensor is mounted in the wall of the
cylindrical chamber with the top end of the chamber open and the bottom end placed
against the skin. As in the closed cup method, the cylindrical chamber defines a volume
of undisturbed air in which the transfer of water vapor is controlled by diffusion in air. A
few minutes after applying the chamber to the skin, the sensor reaches a new equilibrium
in which the rate of flow of vapor out of the open end of the cylinder is equal to the rate
of flow of vapor out of the skin, the TWL. In this condition the rate of flow can be
calculated, according to the principles of diffusion, from the product of diffusion
coefficient of water vapor in the air and the concentration gradient. For the configuration
shown in Figure 9.19(a) the sensor is mounted at a distance l from the end of the cylinder
where the humidity is determined by the room condition. Thus the increase in sensor
response divided by the length determines the humidity gradient. The equation for TWL
measurement with the open cup method is then:
TWL =

(V V0 ) K D
l

(9.36)

where V and V0 are equilibrium and initial sensor voltages, K is an instrument constant,
D is the diffusion coefficient of water in air and l is the distance between the sensor and
the end of the cylindrical chamber open to the room air. Air movement and humidity are
the greatest drawbacks of this method. This method is currently used in commercially
available devices.

35

9.13 Skin
Sensor
voltage

Sensor
Vo
Skin

Time

(a)

(b)

Figure 9.19 The open cup hygrometer: (a) configuration of measurement cylinder, (b) typical
sensor output curve

In the flow hygrometry method, the investigator has the option of selecting the
water content of the sweep gas within the limits of the humidity sensor. For indirect
methods, unless measurements are made in a humidity-controlled room, the variable
ambient condition can have a profound effect on the resulting measurements.
9.13.2 Color

Among the physical skin parameters, color is important in clinical dermatology. The
experienced dermatologist uses color information in several ways. First, he frequently
bases his first-view diagnosis or elaboration of differential diagnosis to a considerable
amount on specific colors of the skin lesions themselves. Second, the color differences
often allow the clinician to judge the distribution of lesions on the human body, even
without close inspection and palpation. Finally, the intensity of color may provide useful
information about the severity of a pathological process, and changes in color to normal
tell the dermatologist if the treatment works.
The color under which we perceive the skin depends on a number of variables,
including pigmentation, blood perfusion, and desquamation (shedding of skin). Because
color perception is a subjective sensory and neurophysiological process, the evaluation of
color is highly observer dependent. This has been a concern not only in dermatology, but
also in industries such as dye production and printing in which highly consistent colors
are necessary. In order to measure color objectively, color-measuring devices have been
developed instead of having it judged by subjective observers.
These devices are useful in dermatology, skin pharmacology, toxicology and
cosmetology because they allow a quantitative measurement instead of a historically
subjective grading of reaction.
Colors may be described by their hue (color position in the color wheel),
lightness (called value) and saturation (called chroma). Any color is expressed as a
combination of (H/V/C) (hue, value, chroma) (Berardesca, 1995).

36

Lung, Kidney, Bone and Skin

Dermaspectrometer and Erythema meter

The instruments used to measure the variations in the color of skin are the
Dermaspectrometer and the Erythema meter. Their working principle can be explained as
follows: Inflammatory skin erythema is the result of an increased presence of
erythrocytes in skin structures. In inflammation, cutaneous vessels are dilated and blood
flow is increased. Because oxyhemoglobin has absorption in the spectral range from 520
to 580 nm, it absorbs green light, while red light is reflected to the absorber. Changes in
skin redness affect the absorption of green light, but affect less that of red light. An
erythema index of the skin can therefore be based on the ratio between the reflection of
red and green light.
Erythema index = log10

( Intensity of reflected red light)

(Intensity of reflected green light)

(9.37)

Similarly melanin pigmentation of the skin lead to an increased absorption both in the
green and the red parts of the spectrum, and a melanin index may be defined as follows:

Melanin Index = log10 (Intensity of reflected red light)

(9.38)

The DermaSpectrometer emits green and red light (568 and 655 nm) from an
LED source. The Erythema Meter emits green and red light (546 and 671 nm) from a
tungsten lamp. Figure 9.20 shows that the light reflected from the skin is detected with a
photosensor. A microprocessor calculates the erythema and the melanin index, which is
then displayed.

9.13 Skin

37

Green
Red

Green

Red

Epidermis

Superficial vascular
plexus

Dermis
Subcutis
Deep vascular plexus

Figure 9.20 The ratio of reflected red to green light is measured in the Dermaspectrometer and
erythema meter.

Chromameter and Micro Color

The Chromameter and Micro Color are another pair of instruments based on the same
working principle as the DermaSpectrometer and Erythema meter. The perceived color of
an object as the spectral distribution of reflected light depends on the light with which the
object is illuminated. This refers both to the color of lighting and the angle under which
the light hits the object. As the perceived color of the object may be expressed as the
proportion of the remitted to the absorbed light, equal lighting over the whole visible
spectrum is essential. Selective lighting results in limited color information; e.g., a red
object would appear black when perceived under a green lamp.
Lighting is achieved using a Xe flash lamp that emits an intensive white light
covering the whole visible spectrum. In order to compensate for variation in illumination,
part of the emitted Xe light is sent to a set of color sensors, whereas the rest illuminates
the object (dual beam system, Figure 9.21(a)). The color sensors analyze the illuminating
light, and through a microprocessor, the light reflected from the object is adjusted for
variations in the illuminating light. Because the measuring unit is a hollow chamber, the
light emitted from the flash lamp hits the surface of the object from all angles. In the
Chromameter, only the light remitted at 0 to the axis of the instrument (90 to the object
surface) is collected for color measurement (d/0 measurement principle, Figure 9.21(b)),
whereas in the Micro Color, the light reflected at 8 to the axis (d/8 measurement) is
collected. The reflected light is transferred to three photodiodes. Before each diode, a

38

Lung, Kidney, Bone and Skin

color filter ensures a specific spectral sensitivity with peaks at 450 nm (blue), 550 nm
(green), and 610 nm (red). These photodiodes simulate the human eye with its three blue,
green, and red sensitive cones in the central fovea. The light reaching the sensors is
transformed into electric signals that are used by the microprocessor for the calculation of
the color values.
Reflected light
Sensor
Sensors
Light
Emitted light
Object

(a)

Object

(b)

Figure 9.21 (a) The spectrum of object light is compared with the spectrum of the light source
alone to allow for differences in lighting of the object (b) Illuminating light is diffuse whereas
measured light is perpendicular to the object (d/0).

9.14 References
Ashman, R. B. 1989. Experimental techniques. In S. C. Cowin (ed.) Bone Mechanics.
Boca Raton FL: CRC Press.
Berardesca, E., Elsner, P., Wilhem, K. -P. and Maibach, H. I. 1995. Bioengineering of the
Skin: Methods and Instrumentation. Boca Raton FL: CRC Press.
Bhattacharya, A. and McGlothlin, J. D. 1996. Occupational Ergonomics: Theory and
Applications. New York: Marcel Dekker.
Black, J. 1988. Orthopaedic Biomaterials in Research and Practice, New York:
Churchill Livingstone.
Bronzino, J. D. (ed.) 2000. The Biomedical Engineering Handbook. 2nd ed. Boca Raton
FL: CRC Press.
Cameron, J. R. and Skofronick, J. G. 1978. Medical Physics. New York: John Wiley &
Sons.
Cohen, S. H. 1981. Non-invasive Measurements of Bone Mass and Their Clinical
Applications. Boca Raton FL: CRC Press.
Dowson, D. 1990. Bio-tribology of natural and replacement synovial joints. In V. C.
Mow, A. Ratcliffe and S. L.-Y. Woo (eds.) Biomechanics of Diarthrodial Joints. Vol.
II. New York: Springer-Verlag.
Ducheyne, P. and Hastings, G. W. 1984. Functional Behavior of Orthopedic
Biomaterials Vol. 1: Fundamentals, Vol. 2: Applications. Boca Raton FL: CRC Press.

9.14 References

39

Engin, A. E. 1990. Kinematics of human shoulder motion. In V. C. Mow, A. Ratcliffe

and S. L.-Y. Woo (eds.) Biomechanics of Diarthrodial Joints. Vol. II. New York:
Springer-Verlag.
Feinberg, B. N. 1986. Applied Clinical Engineering. Englewood Cliffs NJ: Prentice-Hall.
Fung, Y. C. 1993. Biomechanics: Mechanical Properties of Living Tissues. 2nd ed. New
York: Springer.
Gregory, M. C. 1988. Kidney, artificial. In J. G. Webster (ed.) 1988. Encyclopedia of
Medical Devices and Instrumentation. New York, John Wiley and Sons.
Hobbie, R. K. 1997. Intermediate Physics for Medicine and Biology. 3rd ed. New York:
Springer-Verlag
Hologic. 2000. Products. [Online] http://www.hologic.com/
Kryger, M. H. 1994. Monitoring respiratory and cardiac function. In M. H. Kryger, T.
Roth, and W. C. Dement (eds.) Principles and Practices of Sleep Medicine, 2nd ed.
Philadelphia: W. B. Saunders.
Markolf, K. L., Gorek, J. F., Kabo, J. M., Shapiro, M. S. and Finerman, G. A. M. 1990.
New insights into load bearing functions of the anteriaor cruiate ligament. In V. C.
Mow, A. Ratcliffe and S. L.-Y. Woo (eds.) Biomechanics of Diarthrodial Joints. Vol.
I. New York: Springer-Verlag.
Marks, R. and Payne, R. A. (eds.) 1981. Bioengineering and the Skin. Boston: MTP Press
Ltd.
Measurements Group, Raleigh NC. 2000. Interactive guide to strain gage technology.
[Online] http://www.measurementsgroup.com/
Mow, V. C. and Hayes, W. C. 1991. Basic Orthopedic Materials. New York: Raven
Press.
Mow, V. C. and Soslowsky, L. J. 1991. Friction, lubrication, and wear of diathrodial
joints. In V. C. Mow and W. C. Hayes (eds.) Basic Orthopaedic Biomechanics. New
York: Raven Press.
National Institutes of Health. 1994. Bioelectrical impedance analysis in body composition
measurement, Technology Assessment Conference Statement. December 1214. 1
35.
Patterson, R. 1989. Body fluid determinations using multiple impedance measurements.
IEEE Eng. Med. Biol. Magazine, 8 (1): 168.
Petrini, M. F. 1988. Pulmonary function testing. In J. G. Webster (ed.) Encyclopedia of
Medical Devices and Instrumentation. New York: John Wiley and Sons.
Peura, R. A. and Webster, J. G. 1998. Basic sensors and principles. In J. G. Webster (ed.)
Medical Instrumentation: Application and Design. 3rd ed. New York: John Wiley
and Sons.
Primiano, F. P. 1998. Measurements of the respiratory system. In J. G. Webster (ed.)
Medical Instrumentation: Application and Design. 3rd ed. New York: John Wiley
and Sons.
Smallwood, R. H. and Thomas, S. E. 1985. An inexpensive portable monitor for
measuring evaporative water loss. Clin. Phys. Physiol. Meas., 6: 14754.
Webster, J. G. 1988. Encyclopedia of Medical Devices & Instrumentation. New York:
John Wiley and Sons.

40

Lung, Kidney, Bone and Skin

Webster, J. G. 1998. Measurement of flow and volume of blood. In J. G. Webster (ed.).

Medical Instrumentation: Application and Design. 3rd ed. New York: John Wiley
and Sons.
Woo, S. L. -Y., Weiss, J. A. and MacKenna, D. A. 1990. Biomechanics and morphology
of the medical collateral and anterior cruciate ligaments. In V. C. Mow, A. Ratcliffe
and S. L.-Y. Woo (eds.) Biomechanics of Diarthrodial Joints. Vol. I. New York:
Springer-Verlag.
Woo, S. L.-Y. and Young, E. P. 1991. Structure and function of tendons and ligaments.
In V. C. Mow and W. C. Hayes (eds.) Basic Orthopaedic Biomechanics. New York:
Raven Press.
Zhu, W. B. and Mow, V. C. 1990. Viscometric properties of proteoglycan solutions at
physiological concentrations. In V. C. Mow, A. Ratcliffe and S. L.-Y. Woo (eds.)
Biomechanics of Diarthrodial Joints. Vol. I. New York: Springer-Verlag.

9.15 Problems
9.1
9.2

9.3

9.4
9.5
9.6
9.7
9.8
9.9
9.10
9.11

Sketch a spirometer output. Label and describe lung volumes that it can measure.
Label and describe lung volumes that it cannot measure.
You attach a 10 k rotary potentiometer with 5 V across it to a spirometer
pulley so the potentiometer wiper voltage varies from 0 to 5 V. You notice that a
10 k from the wiper to ground causes loading error. Draw an op amp circuit
that prevents this loading error.
You wish to measure lung volume using a body plethysmograph. Sketch the
equipment. Describe (1) sensor type and location, (2) system and method for
calibration, (3) procedure for measurement, (4) results of measurement, (5)
equation, and (6) any corrections required.
In a 3000 L body plethysmograph with closed shutter, the subject blows to
achieve a lung pressure of 10000 Pa. The box pressure changes by 10 Pa.
Calculate the lung volume. State any approximations.
Give the equations and describe how to measure airway resistance.
Given resistivity of the thorax is 300 cm. Assume electrode area and thoracic
dimensions and estimate the impedance of the thorax for use in impedance
plethysmography for measuring pulmonary volume changes.
Describe the importance of measuring pulmonary air-flow. List the different
methods used for the measurement. Explain any one of them.
Sketch a graph of the results obtained when measuring diffusing capacity of the
lung. Label axes and significant quantities.
The measured plasma concentration of creatinine is 1 mg/100 mL. The urine
flow is 4 mL/min. The urinary concentration is 30 mg/100 mL. Calculate the
glomerular filtration rate.
Describe a pyelogram. Explain its importance in diagnosing kidney disease.
Explain how this test is performed clinically.
Describe safety devices required for hemodialysis.

9.15 Problems

9.12
9.13
9.14
9.15

9.16
9.17
9.18

41

Find the transmembrane pressure given KUF is 4 (mL/h)/mmHg. Total fluid to

be removed is 2810 mL and hours on the dialysis is 4 h.
Find the negative pressure across the membrane given transmembrane pressure
is 315 mmHg. The arterial pressure and venous pressure are 130 mmHg and 100
mmHg respectively.
Explain why we measure bone stress. List and describe different sensors that can
be used for this measurement.
A bone specimen with a radius of 1.26 cm is tested in a uniaxial tension test.
The original length of the specimen is 30 cm. After applying a force of 1000 N
the length is changed to 31.5 cm. Calculate the stress and strain generated in the
specimen. Calculate the elastic (Youngs) modulus.
Describe arthrotripsometers, what they measure and their operating principle.
Also explain how there is a exponential decay of the oscillations of the
pendulum with respect to time.
Explain the importance of measuring water loss. Explain how water loss is
measured using flow hygrometry.
Explain how skin color is important to a dermatologist. List different methods of
skin color measurement and explain any one of them.

10.5 Measurement of Body Movement

10
Body Temperature, Heat, Fat, and Movement
Chao-Min Wu

This chapter deals with measurements of physical parameters from the total body, such as
temperature, heat, fat, and movement. These four parameters are somewhat related
because they all deal with energy. Heat is the thermal energy content of a body. It is
defined as the vibratory motion of its component particles. Body heat is the result of a
balance between heat production and heat loss, which occurs through conduction,
convection, radiation and evaporation. Temperature is a measure of the tendency of a
body to transfer heat from or to other bodies. Balancing heat production within the body
against heat loss to the surroundings determines the body temperature. Body fat is energy
related in that it is part of the heat storage process. We discuss how the human body
regulates body temperature and how this regulation is related to heat production, heat loss,
and heat storage in the form of body fat. We further discuss clinical temperature
measurement techniques and common devices. Then we move our discussion to various
methods of body heat and body fat measurements. We compare various techniques
among these measurements as well. Finally, we cover the measurement of body
movement. In each section, we explain why and how to measure these physical
parameters.

10.1

Regulation of Body Temperature

Body temperature is regulated by a well-designed control system. The average normal

body temperature remains almost constant, within 0.6 C for a healthy person. It is
generally considered to be 37 C when measured orally. When speaking of the body
temperature, we usually mean the temperature in the interiorthe core temperature.
Deviation of the core temperature beyond the normal range is a signal of a variety of
diseases and could have a great impact on a person's life if not treated appropriately.
Temperature is an indicator of the health of a person. While the core temperature is
maintained constant, the surface temperature, the temperature of the skin or subcutaneous
tissues, varies with the temperature of the surroundings. This temperature is an important
factor when discussing heat loss through sweating (evaporation) from the skin.

Body Temperature, Heat, Fat, and Movement

Regulation of body temperature uses feedback control. It is analogous to room

temperature control (Milsum, 1966). In room temperature regulation, we preset the
desired temperature on the thermostat as the reference temperature. The thermostat
controls the furnace or air-conditioner (actuator) to increase or decrease the room
temperature to the preset temperature. When the ambient air temperature drops, the
furnace turns on to try and match the preset temperature. During this process,
temperatures from sensors inside the thermostat and from the ambient air are fed back to
the thermostat to speed up the response and allow fine control of room temperature.
Figure 10.1 shows a block diagram of body thermoregulation. In Figure 10.1,
the hypothalamus in the brain is the temperature-regulating center for the body.
Functionally, it is divided into two regions called the heat loss control region and the heat
maintenance region. The heat loss control is located in the anterior hypothalamus, which
contains more warm-sensitive neurons than cold-sensitive ones. Temperature-sensitive
neurons act like the temperature sensor in the thermostat. It is clear that the set-point
temperature for a healthy person is 37 C. All the temperature control mechanisms
attempt to bring the body temperature back to this normal level. However, low skin
temperature and the presence of substances released from toxic bacterial cell membranes,
such as lipopolysaccharide toxins, cause the critical set-point temperature to rise. These
substances are called pyrogens and cause fever during the period of illness (Guyton and
Hall, 1996). This change of the set-point level acts as a unit step input to the body
temperature control system, which is considered a first-order system. The time constant
for our body to match this set-point change is about 3 h.

10.1 Regulation of Body Temperature

Control center

Actuators

Anterior
hypothalamus
set-point

Sweating
Skin vasodilation
Decrease metabolism

Posterior
hypothalmus

Muscle shivering
Skin vasoconstriction
Increase metabolism

Core
temperature
Skin
temperature
+

Skin thermal
receptors

Hypothalamic thermal
receptors

Figure 10.1 The human temperature regulation system can increase or decrease body temperature.

The heat maintenance region is located in the posterior hypothalamus where it

combines sensory signals transmitted from the anterior hypothalamus and peripheral
temperature sensors to control the temperature-increasing or temperature-decreasing
mechanisms of the body. This part of the hypothalamus acts like a thermostat controller
in room temperature regulation. When the body is too hot, the heat maintenance region
triggers temperature-decreasing procedures to reduce body heat by sweating, skin
vasodilation, and decreasing tissue metabolism. Sweating is the main mechanism by
which the body causes heat loss via water evaporation. The effect of skin vasodilation is
an increase in heat loss by increasing blood flow to the skin surface. Vasodilation is
handled through the autonomic nervous system. Excessive tissue metabolism is inhibited
to decrease heat production. In turn, it decreases body temperature.
On the other hand, temperature-increasing procedures are instituted when the
body temperature is too cold. They are: skin vasoconstriction, shivering, and increased
thyroxine secretion. The reaction of skin vasoconstriction, in contrast to skin vasodilation,
is caused by stimulation of the posterior hypothalamus. Shivering is a reaction of skeletal
muscles to call for an increase in heat production of the body. It is excited by the cold
signals from the skin receptors (Guyton and Hall, 1996). The primary motor center for
shivering is located at the posterior hypothalamus (dorsomedial portion). The response
time of skin vasomotor control and shivering is within minutes. However, the increase of
thyroxine secretion is a long-term process of heat production increase through the

Body Temperature, Heat, Fat, and Movement

endocrine system. It usually takes several weeks to complete. Thyroxine is the hormone
secreted by the thyroid gland to increase the rate of tissue metabolism.
In addition to the physiological control of body temperature, behavioral
response to the change of environment plays an important role in body temperature
control, especially in a very cold environment. It is usually an efficient way to maintain
our body temperature. For example, we put on more clothes when we expect severe cold
weather.

10.2 Clinical Temperature Measurement

There are two kinds of clinical temperature measurements: surface or skin temperature
measurement and core temperature measurement. We measure temperature with
thermometers. A thermometer is a device that detects a change in temperature. This is
done by detecting a change in the physical property of a sensor. This physical property
can be a mechanical expansion of a temperature dependent liquid, such as mercury, or an
electric resistance change of a thermally sensitive resistor (thermistor). The thermometer
uses these materials as sensors to measure temperature. Traditionally, we measure
temperature from an oral cavity, rectum, or under the arm by contact measurements. The
measured temperature varies from site to site, ranging from the oral temperature, which is
lower than the core temperature by about 0.5 C to the rectal temperature that is lower
than the core temperature by as much as 2 C (Fraden, 1991). These temperature
measurements need patient cooperation and are affected by the presence of food, drink,
and physical activity.
The tympanic membrane that connects the outer ear (ear canal) and the middle
ear (three small auditory bones) provides an ideal location for temperature measurement.
Theoretically, temperature measurement taken near the tympanic membrane not only is
close to the core temperature (proximate to the hypothalamus), but also is not affected by
interfering factors. In practice, noncontact measurement of the tympanic membrane
temperature is not superior to other measuring sites in reflecting the real core temperature.
The measured temperature is still lower than the core temperature by about 0.5 to 1 C. It
all depends on how close the measurement is made to the tympanic membrane. However,
it provides a convenient and comfortable site for reliable and accurate temperature
measurement for a variety of patients.
10.2.1 Surface temperature measurement
Radiation from the skin surface is the main mechanism of heat loss from the body. It
accounts for 60% of the body heat loss under normal conditions (Guyton and Hall, 1996).
The relationship between the energy emitted from the skin surface and the skin surface
temperature forms the basis of surface temperature measurement. The surface
temperature is determined by the skin blood supply and the condition of subcutaneous
tissues. It is a good indicator for bone fractures and inflammation. Abdominal placement
of a thermal sensor (thermistor) for skin temperature measurement is important in
neonatal monitoring against cold stress. Medical thermography is a technique to scan the
skin surface and map its thermal distribution. Physicians have used this technique to

10.2 Clinical Temperature Measurement

diagnosis tumors and breast cancers. However, modern imaging techniques, such as MRI
and mammography, are more popular diagnostic tools for these carcinomas.
Liquid crystal thermometer
Liquid crystal thermometers are made of chemical compounds with the property of
reflecting light within a range of temperature (26 to 40 C). For example, the crystals can
be formulated to a change in color through red, yellow, green, and blue as the substance
is heated from 38 C through to 40 C. They are constructed by impregnating Mylar
strips with encapsulated liquid crystals. With attached adhesive, they are often used for
surface temperature measurements, especially when monitoring the temperature trend in
an operation room. Liquid crystal thermometers are easy to use, inexpensive, and
disposable but less precise than other thermometers (see next section). In general, their
resolution is around 0.2 C to 0.5 C.
10.2.2 Core temperature measurement
While surface temperature measurement is used by physicians as a diagnostic tool for
specific diseases, core temperature measurement is a rather routine method used in
hospitals, clinics, or even at home. It can detect fever caused by pyrogens that are
released from virus, bacteria, and degeneration of old tissues. High and lasting fever
above 41 C can cause damage to the brain and other internal organs that are vital to
human life. Therefore, core temperature measurement should be accurate and fast.
Temperature measurement varies from site to site. While Bair and Davies (1995) reported
that temperature measurement at the pulmonary artery reflects the best core temperature
in critically ill patient monitoring, rectal or tympanic temperature measurement is also
quite reliable and has less danger, less cost, and more accessibility. This section discusses
basic principles, advantages, and disadvantages of different types of thermometers for
core temperature measurements.
Mercury thermometer
The mercury thermometer is a mechanical thermometer. It is based on the principle of
volume expansion. The temperature-dependent mercury expands when the temperature
increases. The result of the change can be read on a calibrated scale. The mercury
thermometer is the most commonly used device to measure body temperature at home
because of its low cost and ease of use. Its accuracy varies widely depending on how well
the expansion of mercury is calibrated. The time to measure the temperature is in the
range of 3 to 5 min. The cool mercury pulls down the temperature of the tongue. Then it
takes 3 to 5 min for new perfused blood to warm up the mercury. It is affected by how
well the probe and tissue surface are coupled. Infection and crosscontamination are its
main disadvantages, although constant decontamination of the thermometer can reduce
the risk of infection. Other disadvantages are a potential damage to mucosal tissues when
measured orally or rectally, and possible mercury intoxication due to the broken glass.

Body Temperature, Heat, Fat, and Movement

Caution should be taken in handling and storing of the mercury thermometer. The
accuracy of the mercury thermometer is usually manufacturer-dependent.
Electronic thermometer
Thermometers that use temperature-responsive sensors are called electronic
thermometers. Because their sensor mass is smaller and they use electronic circuits, their
response time is faster than that of mercury thermometers. Sensors may vary their electric
resistance or voltage. Thermistors are thermally sensitive resistors with either negative
(NTC) or positive (PTC) temperature coefficients. The electric resistance of a NTC type
thermistor decreases as the temperature increases. Figure 10.2 shows examples of
resistancetemperature curves for NTC type thermistors. They are composed of
dissimilar materials with temperature-dependent electrical resistance, such as oxides of
nickel, manganese, or cobalt metal, to enhance their temperature sensitivity. Dissimilar
materials are mixed homogeneously and produced in the form of beads, rods, disks, and
other shapes.
3

10

= 3500 K
= 4000 K

R/R25C

10

= 3000 K

10

10
50

50
100
Temperature (C)

150

200

Figure 10.2 Examples of resistancetemperature curve for three NTC thermistors, = 3000 ,
3500 K, and 4000 K.

The resistancetemperature curve shown in Figure 10.2 is based on Eq. (10.1).

The resistance of the thermistor at temperature T in degrees kelvin (K) is given as

10.2 Clinical Temperature Measurement

1 1
RT = R0 exp

T T0

(10.1)

where is the characteristic temperature of the material, R0 is the resistance at T0 = 298

K. The parameter is temperature dependent, however, it can be considered as
temperature independent for medical applications because the range is only 37 5 C.
The temperature coefficient, or sensitivity, , can be defined by differentiating
Eq. (10.1), which yields

dR
= R T T = 2 .
(10.2)
dT
T

where is temperature dependent and exhibits the nonlinearity of thermistors. At 27 C

(300 K), given = 4000 K, is 4.4%/K.
In contrast to the homogeneous composite of dissimilar materials for the
thermistor, a thermocouple is composed of dissimilar materials that are constructed as
fused junctions of two materials (e.g., copper and constantan wires). Bare wires have a
rapid response time. Wires contacting a metallic sheath have a slower response time.
Wires enclosed in an insulator have the slowest response time. When the thermocouple is
heated, a current flows from the measuring side (hot junction) to the reference side (cold
junction). It develops an electric potential, which can be measured when we place a
voltmeter between open ends as shown in Figure 10.3. In 1823, Seebeck first reported
this phenomenon, called the Seebeck effect. Figure 10.3 shows the measuring principle of
a J type thermocouple. An approximate formula for the Seebeck voltage that is related to
junction temperatures T1 and T2 is

V = C1 (T1 T2 ) + C 2 T12 T22

(10.3)

where C1 and C2 are constants that depend on the thermocouple pair with T in kelvins. If
the temperature difference T1 T2 is small, the second term of Eq. (10.3) can be dropped.
The sensitivity or thermoelectric power of the thermocouple is the derivative of Eq. (10.3)
with respect to T1 is

dV
dT1 .

(10.4)

For example, the sensitivity for the copperconstantan thermocouple is 45 V/C at

20 C (Cobbold, 1974). To prevent changes in T1 T2 due to ambient temperature
change, an electronic cold junction is used.

Body Temperature, Heat, Fat, and Movement

Voltmeter
Copper

Copper
Constantan

Cold junction, T2
(Reference)

T1 , Hot junction
(Measuring probe)

Figure 10.3 The J type thermocouple is formed from copper and constantan.

Electronic thermometers use thermistors or thermocouples as sensors to measure

temperature in the mouth, rectum, or axilla (armpit). Thermistor type sensors are suitable
for continuous temperature monitoring in patients during surgery and for infants' skin
temperature monitoring on the abdomen. Microprocessor-based electronic thermometers
use either the steady-state or predictive method to improve the response time for
temperature measurement to less than 1 min. A thermometer with steady-state design
monitors the temperature reading until it reaches a thermal steady state between the probe
and measuring site. Once it is in the steady state, the display shows the temperature of the
sensor In contrast, a thermometer of the predictive type measures the temperature based
on a manufacturer's algorithm that is developed from several experiments of the rate of
rise in temperature to estimate the final temperature (Bair and Davies, 1995). Figure 10.4
shows a simplified block diagram for an electronic thermometer.

Digital
display
Signal conditioner
and final temperature
detector
(processor)

Probe
(thermistors)
Wheatstone
bridge

Audible
alarm

Figure 10.4 An electronic thermometer uses a thermistor sensor probe.

The major advantages of electronic thermometers are the small size of sensors,
ease of use, and fast response time. Their disadvantages are similar to those of mercury
thermometers, such as mucosal damage and crossinfection if not carefully handled. A

10.2 Clinical Temperature Measurement

disposable probe cover is used to prevent cross-infection. Sliding the probe from the floor
of the mouth to the sublingual pocket warms up the probe and decreases oral temperature
measurement error and response time.
Infrared thermometer

A basic infrared thermometer (IRT) design includes four parts: a waveguide (tube) to
collect the energy emitted by the target, a sensor to convert the energy to an electric
signal, an emissivity adjustment to match IRT calibration to the emitting characteristics
of the object being measured, and a sensor temperature compensation circuit to ensure
that temperature variations within the IRT are not transferred to the final output. Figure
10.5 shows a block diagram of an infrared thermometer. The tympanic membrane emits
infrared energy. The collection of infrared radiation is enhanced by a waveguide, which
improves reflection of the infrared radiation onto the sensor when the shutter opens. The
pyroelectric sensor converts infrared radiation into electric signals, which are fed through
the amplifier, multiplexer (MUX), and analog-to-digital converter (ADC) to the
microprocessor for processing and display. The microprocessor handles emissivity
adjustment and temperature compensation, and calculates the patient (target) temperature
according to Eq. (10.5):

Tb =

[ (N

N T0 ) + T04
4

(10.5)

where Tb is the patient temperature, a constant (= 1/Aa), T0 is the reference

temperature (= 310 K), NT0 is the net flux of the reference temperature, and NT is the net
flux of infrared radiation, which is shown as follows.

N T = A a Tb4 Ta4 ,

(10.6)

where A is the effective body (target) area, the StefanBoltzmann constant (= 5.67
108 W/m2K4), a the emissivity of surroundings (sensor), Tb the body temperature,
and Ta the sensor temperature (Ta and Tb are both in kelvins).

10

Body Temperature, Heat, Fat, and Movement

Shutter
Ear
IR

Ambient sensor
Ta
Tb

Sensor
Waveguide

Window

MUX

A/D

Micro
processor

Amp.
Shutter
switch

Digital
display

Figure 10.5 The infrared thermometer opens a shutter to expose the sensor to radiation from the
ear.

The ambient sensor that monitors the sensor temperature (Ta), is a thermistor.
The infrared sensor is a pyroelectric sensor followed by a current-to-voltage converter
that is used for the pyroelectric sensor (Fraden, 1991). The pyroelectric sensor, like the
piezoelectric sensor does not respond to steady inputs and in response to a step input
yields a high-pass first order output of the form e-t/. The whole measurement cycle,
from shutter opening to display, is 1 s. Its accuracy and resolution are about 0.1 C. The
main advantages of the infrared thermometer are that it only contacts the ear canal and
not the eardrum (i.e. less infection), ease of use, and fast response. It is suitable for the
temperature measurement of young or high-risk patients. However, inaccurate
measurements occur when patients have curved ear canals or the cerumen (ear wax)
obstructs the tympanic membrane.

10.3 Measurement of Body Heat: Calorimetry

Calorimetry is the measurement of heat given off by a subject. The term direct
calorimetry is used for the direct measurement of heat loss from the subject by radiation,
convection, conduction, and evaporation. Indirect calorimetry measures respiratory gas
exchange to infer the amount of heat production. Heat production is calculated from
oxygen (O2) consumption and carbon dioxide (CO2) production and converted to energy
expenditure based on an empirical equation. Calorimetry provides the basis for the
science of nutrition. It is used to estimate nutritional requirements of humans and to
evaluate different foods. Calorimetry is also a powerful research tool for energy balance
and thermal physiology studies. Finally, it is used clinically as a diagnostic tool for the
investigation of metabolic disorders, energy requirements of major illness, and sports
medicine. In this section, we will deal with the principles and applications of direct and
indirect calorimetry.

10.3 Measurement of Body Heat: Calorimetry

10.3.1

11

Direct calorimetry

Direct calorimetry has been used on animals and in humans for energy and thermal
regulation studies. Since the beginning of the century, many direct calorimeters have
been built. They are usually classified under one of four principles of operation: gradient
layer calorimeter, airflow calorimeter, water flow calorimeter, and compensating heater
calorimeter. The gradient layer calorimeter is usually used for studies in temperature
regulation of human subjects while they are performing different work. The airflow
calorimeter is popular in studies of energy expenditure for humans and animals. The
water flow calorimeter can be used for both applications without social isolation. The
compensating heater calorimeter is commonly used in experiments for small animals.
Webb (1985) provides detailed information on the historical development of these direct
calorimeters and a list of human calorimeters.
Gradient layer calorimeter

Gradient layer calorimeters are isothermal calorimeters. They are chambers lined on all
surfaces with a layer of insulating material and surrounded by a heat sink metal wall or a
constant temperature water jacket. As sensible (i.e. nonevaporative) heat from the subject
passes through the thin layer of insulating material to the heat sink (metal wall or water
jacket), the temperature gradient across the layer of insulation is sensed by a network of
thermocouples (e.g. copperconstantan). Benzinger and Kitzinger (1949) first introduced
this type of calorimeter. These thermocouples are used to detect the temperatures of the
inner and outer surfaces. The evaporative heat loss from the subject can be estimated
from the increased moisture in the ventilating air that exits the chamber. The main
objective of the gradient layer calorimeter design is to prevent any sensible heat loss from
convecting out of the chamber by heating up the ventilating air. The design and
construction of a gradient layer calorimeter is a complicated engineering task. It includes
the control of the ventilation and cooling system and instrumentation for the temperature
measurement for a huge chamber. The performance of the calorimeter depends on initial
calibration and routine calibration with known electric energy and is controlled by a
computer. McLean and Tobin (1987) provide a detailed description of how to design and
construct a gradient layer calorimeter. They reported that the accuracy of total heat loss
measurement is around 1% or 3 W within the operation range (10 to 40 C). The
response time of the measurement to a unit change of heat output (i.e. when a person is
first placed inside) is 80% of the change is recorded within 3 min and 90% within 10 min
but the full response takes nearly an hour. Their calibration equations are listed as follows:
VC =

(QD N QD C QD L )
VT =

kc
QD

VDk t

(10.7)
(10.8)

12

Body Temperature, Heat, Fat, and Movement

C vC C vC

QE C + QC C + p
V
V

VE =
ke
QC = (V + V )k + QC
N

VC
QC E = (VE VT )k p
vC

(10.9)
(10.10)
(10.11)

where VC, VT, and VE are measured voltages from the chamber, thermocouples, and
evaporative heat, respectively; QD N and QD E are the rates of nonevaporative and
evaporative heat production in the chamber; QC C is the rate of nonevaporative heat
convection out of the chamber; QD L is the rate of heat leakage through any gaps in the
heat-sensitive layer; VD and D are chamber and evaporative heat measuring air flow rates;
kc, ke, and kt are constants (sensitivities); p is a term due to unequal sensitivities or
unequal sampling rates; during the calibration, as performed with the chamber controlled
at 25 C and the evaporative heat measuring unit at 10 C, the desired value k = kc = ke
VD /D = kt VD = 46.5 W/mV (40 kcal/hmV) is obtained by adjusting the output circuit of
the calorimeter when the ventilation system is in operation and with no load in the
chamber ( QC N = QD E = 0).
The advantages of the gradient layer calorimeter are accuracy and fast response;
however, they are expensive and not transportable. Figure 10.6 shows that they are
complicated in design and construction.

10.3 Measurement of Body Heat: Calorimetry

13

Insulation layer

Constantan

Metal wall
Cooling tubes
Copper

Thermocouple
connection

Figure 10.6 In a gradient layer calorimeter, thermocouples measure the difference in temperature
across the wall. Ventilating system and measurements not shown.

Air flow calorimeter

An airflow calorimeter is a convection calorimeter where heat loss through the surface
is prevented by insulation. Heat loss from the subject is carried away from the chamber
by airflow. The rate of heat removal by the airflow is estimated from the mass flow
rate and temperature rise of the air. An important objective for the design of an airflow
calorimeter is to prevent any water condensation on the surface of the chamber. Figure
10.7 shows the principle of an airflow calorimeter. The seated subject is confined in a
chamber with polyurethane insulation applied to the outsides. The heat loss from the
subject is transferred to the ventilating air, whose temperature is measured at the inlet and
the outlet of the chamber. The heat loss ( Q a ) from the subject is the product of air mass
flow rate ( m ), specific heat (ca), and temperature change (T2 T1 ) of the ventilating air:
Q a = m c a (T2 T1 )

(10.12)

14

Body Temperature, Heat, Fat, and Movement

Thermometer

Air out

T2
Air in

Thermometer
Insulated chamber
(polyurethane)
T1

Figure 10.7 An air flow calorimeter measures inlet and outlet temperatures, flows, and humidity.

The advantages of the airflow calorimeter are simplicity of construction and

fast response to heat change (7 to 20 min). However, they need very accurate humidity
and temperature measurements.
Water flow calorimeter

A water flow calorimeter is a heat-sink calorimeter. Similar to a convection calorimeter,

the water flow calorimeter uses, instead of air, water as the heat exchanger. The removal
rate of heat is estimated from the flow rate and temperature rise of the coolant (water).
Figure 10.8 shows a suit calorimeter, which is a typical water flow calorimeter. Webb
(1985) gives a detailed description of their suit calorimeter. Their suit calorimeter
consists of a water-cooled undergarment, insulating outer garments, a controlled heater
and the cooling machinery for the water loop, and the related devices for control and data
handling. The water-cooled undergarment includes a garment made of washable synthetic
fibers, an elastic mesh material, and an array of polyvinyl chloride tubes in contact with
the skin. Each 20% of the total of small tubes (a total length of 80 m with an inner
diameter of 1.6 mm) covers the head and neck, torso, each leg, and both arms,
respectively. The circulating water is pumped, at 1.5 kg/min into the inlet tube of the
garment at the ankles, wrists, and neck and collected from the outlet tubes at the waist
and neck. The subject's heat loss to the water ( Q w ) is estimated from the mass flow rate
( m ), specific heat (cw) and temperature difference between water outlet (T2) and inlet
(T1). The formula is similar to Eq. (10.12). The advantages of the water flow calorimeter
are its fast response (30 s) and it is fully transportable. In addition, the subject is not
socially isolated and confined in the chamber or room. One main disadvantage reported
by Webb (1985) is that the heat loss from the subject occurs by conduction, from skin to
the cooled tubing. This method of heat loss is physiologically unusual.

10.3 Measurement of Body Heat: Calorimetry

Water out

15

T2

Polyvinyl chloride tubes

Outlet thermometer
T1
Water in

Inlet thermometer
Figure 10.8 The water flow calorimeter measures the inlet and outlet water temperature.

Compensating heater calorimeter

A compensating heater calorimeter supplies electric heater power to maintain a constant

temperature inside an uninsulated chamber. The heater is needed to protect against heat
loss to the environment. Figure 10.9 shows that when the subject's heat is added to the
chamber, the heater needs less power to maintain the constant temperature of the chamber.
The decreased power then can be used to calculate the subject's heat loss. The advantages
of this type of calorimeter lie in its speed of response (on the order of 1 min) and
lightweight construction. It is suitable for small caged animals and no longer used for
human subjects.
Thermostat
Feedback control

AC
Power
source

Power
meter

Heater
Figure 10.9 The compensating heater calorimeter requires less heater power when the subject
supplies heat.

16

Body Temperature, Heat, Fat, and Movement

10.3.2 Indirect calorimetry

Indirect calorimetry estimates heat production by quantitative measurements of the

oxidative processes of substrates, especially the measurement of the rates of oxygen
consumption ( VDO 2 ) and carbon dioxide production ( VDCO 2 ). Indirect calorimetry
constitutes a noninvasive and nonintrusive technique that can be applied in basic research,
clinical and field studies of energy requirements for nutritional support in patient care,
and sports medicine. We discuss three commonly used methods of indirect calorimetry in
this section: open-circuit systems, closed-circuit systems, and the double-labeled water
method. The closed-circuit system and the open-circuit system are further grouped into
the respiratory gas exchange method. Most ambulatory, laboratory, and bedside uses for
the estimation of energy expenditure are based on measurements of respiratory gas
exchange. Respiratory exchange methods depend mainly on the measurement of oxygen
consumption ( VDO 2 ), either alone or combined with measurement of carbon dioxide
( VDCO 2 ) and methane production ( VDCH 4 ), and sometimes urinary nitrogen excretion
( ND ). The apparatus that collects gas includes a mouthpiece with a nose clip, a ventilated
hood or bed canopy, or a whole body chamber. The whole body chamber system
sometimes is considered as a new category. Because the whole body chamber is also a
closed-circuit system but conducted in a confined environment, we consider it as one of
the closed-circuit systems. The differences among these methods are in response time,
accuracy, duration of measurement, and the degree of confinement. In general, the whole
body chamber system is used for physiological research over long periods of time (24 h
or longer). The double-labeled water method allows a free living environment but
requires expensive isotopes for the measurement. A detailed review of techniques for the
measurement of human energy expenditure and commercially available product
information are given by Murgatroyd et al. (1993) and Branson et al. (1995).
Open-circuit system

The open-circuit system determines VDO 2 from the minute ventilation rate ( VDE ) and the
difference ( F ) between inspired ( FI ) and expired ( FE ) gas concentrations. This
calculation of VDO 2 includes a correction for change in barometric pressure, temperature,
and relative humidity of the inspired air for adjustment of the volume to standard
temperature, pressure, and dry air condition (STPD: 0 C, 760 mmHg, dry). Energy
expenditure (EE, kcal/day) is estimated by the standard (Weir, 1949) equation:

[(

EE = VDO 2 (3.941) + VDCO 2 (1.11) 1440

(10.13)

where VDO 2 and VDCO 2 are expressed in L/min, and 1440 = the number of minutes in a
day. It assumes that 12.3% of the total calories arise from protein metabolism. When
urinary nitrogen excretion ( ND ) is measured, a third term ( ND (2.17)) in g/min is added to
the bracket of Eq. (10.13) to account for protein correction of energy expenditure.
Open-circuit calorimeters use either the mixing chamber or the dilution
technique to measure inspired and expired concentrations of oxygen ( FIO 2 , FEO 2 ) and

10.3 Measurement of Body Heat: Calorimetry

17

carbon dioxide ( FICO 2 , FECO 2 ), and inspired and expired minute ventilation ( VDI ,
VDE ). In an ideal situation, VDO 2 is determined by
VDO 2 = VDI (FIO 2 ) VDE (FEO 2 )

(10.14)

and VDCO 2 by a similar equation:

VDCO 2 = VDE (FECO 2 ) VDI (FICO 2 ).

(10.15)

However, it is technically difficult to measure the difference between VDI and VDE . The
relationship between VDI and VDE is commonly calculated by equating the quantity of
nitrogen in inlet and outlet air stream, i.e. the Haldane transformation. Then VDI is
obtained by the equation
FEN 2 D
VDI =
VE
FIN 2

(10.16)

where FEN2 = expired nitrogen concentration and FIN2 = inspired nitrogen concentration.
If only N2, O2, and CO2 are analyzed, VI can be obtained by
1 FEO 2 FECO 2 D
VDI =
VE
1 FIO 2 FICO 2

(10.17)

where FEN2 = 1 FEO2 FECO2 and FIN2 = 1 FIO2 FICO2. Substituting Eq.
(10.17) into Eq. (10.14), the formula for VO 2 becomes
(FIO 2 FEO 2 FIO 2 FECO 2 FICO 2 FEO 2 ) 
VO 2 =
VE
1 FIO 2 FICO 2

or
(1 FEO 2 FECO 2 ) FIO 2

FEO 2 VE
VO 2 =
1 FIO 2

(10.18)

when FIN2 = 1 FIO2, because FICO2 = 0.03% in most situations. This equation is used
by most of commercially available systems. In general VDI and VDE are of similar
magnitude but not equal; the same is true for FIO2 and FEO2. The problem with using
the Haldane transformation is that it becomes increasingly inaccurate as the FIO2
increases above 0.6 (see below). Consequently, extreme precision in estimates of these
four variables would be necessary if VDO 2 were to be calculated directly from Eq. (10.14).
If we assume VDI is equal to VDE , there will be significant errors in VDO 2 measurement, for
example in a patient whose respiratory rate is 10 breaths per minute at a 1000 mL tidal

18

Body Temperature, Heat, Fat, and Movement

volume ( VDT ) and whose VDO 2 is 250 mL/min. The respiratory quotient (RQ, given by the
ratio VDCO 2 / VDO 2 ) of this patient is 0.7. This means that 25 mL per breath of O2
produces 17.5 mL per breath of CO2. In this case, the difference between VDI and VDE is
75 mL. If VDO 2 is measured at an O2 concentration of 60% (FIO2 = 0.6) assuming that
VDI is equal to VDE , an error of 18% would be obtained for VDO 2 measurement (205
mL/min). When RQ reaches 1.0, this error becomes small. Measurement of VDCO 2 is
much simpler. When inspired air is room air, it contains only very small amount of CO2
(< 0.03%). VDCO 2 may be calculated accurately from a simpler form
VDCO 2 = VDE (FECO 2 ).

(10.19)

This simpler form is also applied to measurement of VDCO 2 for patients with mechanical
ventilation where FICO2 is 0.0.
Most commercially available open-circuit systems use mixing chambers as in
Figure 10.10. Expired air from the patient passes through a facemask or mouthpiece and
is directed to the mixing chamber. At the end of the mixing chamber, a small sample is
drawn by a vacuum pump through O2 and CO2 sensors that measure FEO2 and FECO2.
A pressure transducer and thermistor are necessary for pressure- and temperaturecompensated gas measurements. A volume transducer is used to measure minute
ventilation ( VDE ). A microprocessor controls the instrument and provides minute-byminute data on O2 consumption and CO2 production, RQ, and energy expenditure of the
subject.
From
Inspired
gas source

O2
Switch

CO2

Computer
control

Pressure
transducer

Vacuum
pump
Thermistor
Exhaust
Patient
Expired gas
from patient

Flow
transducer

Mixing chamber
Figure 10.10 A microcomputer-based open-circuit system includes a mixing chamber, O2 and CO2
analyzers, and the various variables (pressure, flow, temperature) used to calculate VDO 2 .

When measuring respiratory exchange gas, there are two broadly used methods:
the air may be directly expired through a mask or mouthpiece as mentioned above or it
may be drawn from a hood or canopy enclosing the subject's head. The latter is
sometimes called a flow-by system. It needs a pump to draw constant air over a subject's

10.3 Measurement of Body Heat: Calorimetry

19

head and this airflow is mixed with the patient's expired air. The process of VDO 2 and
VCO 2 measurements is the same as before. However, the patient's expired air is diluted
by the constant air flow, the difference in O2 and CO2 concentrations are smaller than
those measured by the former method. Consequently this demands that the analyzers be
more precise and free of drift. The Deltatrac metabolic monitor (DatexEngstrom
Division, Instrumentarium Corp., PO Box 900, FIN-00031, Helsinki, Finland) and
Vmax29n (SensorMedics Corp., Yorba Linda, CA) represent this type of system. It is
suitable for use with a mechanical ventilator or a canopy for patients of all ages. The
Deltatrac consists of an infrared CO2 analyzer, a paramagnetic O2 analyzer, a 4 L mixing
chamber, and a CRT display (Weissman et al., 1990). When the Deltatrac is used with a
mechanical ventilator (Figure 10.11), the expired gases enter the mixing chamber, from
which gas is sampled and analyzed to determine FEO2 and FECO2. After the expired
gases pass through the mixing chamber, the gases are diluted with room air driven by a
flow generator (Q, 40 L/min). VCO 2 is then calculated as

VDCO 2 = Q F * CO 2

(10.20)

where F*CO2 is the concentration of CO2 in the diluted gas sample. The respiratory
quotient (RQ) is then calculated using the Haldane transformation.
RQ =

1 FIO 2
FIO 2 FEO 2
FIO 2
FECO 2

(10.21)

where FIO2 is the O2 concentration from the inspired limb of the ventilator. VDO 2 is then
determined by the formula

VDCO 2
VDO 2 =
.
RQ

(10.22)

20

Body Temperature, Heat, Fat, and Movement

From inspired gas source
Metabolic
monitor
Inspiratory
limb

Humidifier

FIO2
Patient

FECO2,
FEO2
Expiratory limb
Exhalation
valve

CO2 and O2
analyzers

F*CO2
Constant
flow
generator

Mechanical
ventilator
Room air
Figure 10.11 A dilution system used by Deltatrac uses mechanical ventilation. FIO2 is the inspired
oxygen concentration, FECO2 is the true expired carbon dioxide concentration, FEO2 is the
expired oxygen concentration, and F*CO2 is the diluted carbon dioxide concentration.

A microprocessor controls the sampling of gases. CO2 measurements are made

by alternately sampling the diluted CO2 (F*CO2) and the true CO2 concentration
(FECO2). O2 measurements occur by alternately sampling the difference between
inspired and expired O2 (FIO2 FEO2) and the measurement of FIO2. Weissman and
colleagues (1990) reported that the accuracy of the Deltatrac for VDO 2 and
VDCO 2 measurements is within 7%. They also showed a different calculation procedure
for canopy measurement of VDO 2 and VDCO 2 in spontaneously breathing patients.
In addition to the mixing chamber and the dilution techniques mentioned above,
some mixing chamber systems use a breath-by-breath technique for VDO 2 and
VDCO 2 measurements. The breath-by-breath technique was originally designed for
exercise testing and has been adapted for use with a mechanical ventilator. It makes
measurements for every breath of the patient's expiration and differs from the mixing
chamber system in that the mixing chamber is absent. Gas samples are collected on a
breath-by-breath basis and are averaged over a preset time to obtain mixed expired values
(Branson et al., 1995). Furthermore, bedside mixing chamber systems can be modified
and used for ambulatory and field studies by dispensing with the CO2 analyzer to
simplify the procedure. However, this results in lower accuracy.
Closed-circuit system

10.3 Measurement of Body Heat: Calorimetry

21

Closed-circuit systems calculate VDO 2 by measuring over time the volumetric change
from a source of oxygen. They use the volumetric loss and replenishing techniques to
measure VDO 2 . The former observes how long it takes to consume all the oxygen that is
within a spirometer. The latter measures the amount of oxygen needed to keep the
spirometer full. A generic volumetric loss closed-circuit system consists of a CO2
absorber, a breathing circuit, a spirometer (see Figure 9.2) full of 100% O2, and a CO2
analyzer (Figure 10.12). Closed-circuit systems allow VDO 2 and VDCO 2 measurements
during mechanical ventilation and during spontaneously breathing ventilation with a
mouthpiece plus nose-clip or facemask. During measurements, the patient inspires O2
from the spirometer and the expired gases enter the mixing chamber. Samples of the
expired gas are analyzed to obtain FECO2 before passing through the soda lime scrubber.
Movements of the spirometer are recorded as VT and VDE . The volume change in the
spirometer as a function of time is used to calculate the rate of O2 consumption. In a
similar way as previously described, the rate of CO2 production is determined from
measurements of VDE and FECO2. During mechanical ventilation, the bellows in Figure
10.12 is driven by the ventilator to deliver gas to the patient.
From ventilator

Bellows

Spirometer
Patient
CO2
Scrubber As oxygen
supply and
analyzer

Mixing
chamber
Sample
gas

CO2 analyzer

Figure 10.12 A closed-circuit system uses the volumetric loss principle. The spirometer is used as
an oxygen supply and the volume change as a function of time in the spirometer is used to calculate
the rate of oxygen consumption.

The whole body chamber is an early design of a volume-replenishing closedcircuit system. In the whole body chamber the subject is kept in a sealed chamber that is
ventilated with a constant, measured supply of fresh air. The respiratory gas exchanges
produced by the subject are measured in terms of their effects on the composition of
chamber air. Samples of well-mixed air are drawn from the chamber for continuous
analysis. The differences in oxygen and carbon dioxide concentrations in the in-going
and out-going air are measured and used to calculate energy expenditure. McLean and
Tobin (1987) stated that the Atwater and Benedict respiration chamber represents one of
the earliest designs of this type of calorimeter. The fundamental design of a whole body

22

Body Temperature, Heat, Fat, and Movement

chamber is similar to and functions as the mixing chamber in Figure 10.12. During the
experiment, air is drawn from the chamber by an air pump and passes through an external
circuit. This external circuit includes one CO2 absorber, two moisture absorbers, a
spirometer, and an oxygen supply. Moisture in the air is absorbed by passage through a
container of sulfuric acid and this is followed by absorption of CO2 in soda lime; the
moisture given off from the soda lime is then absorbed by a second container of sulfuric
acid. The pressure drop in the system, caused by the uptake of oxygen by the subject and
the absorption of any expired CO2, is detected by the spirometer. As pressure in the
chamber decreases, the volume of spirometer decreases. This then releases a valve on the
oxygen inlet line. As the oxygen supply replenishes the chamber with oxygen, the
spirometer bell rises and the valve is closed again. A meter is used to measure the
delivery of oxygen. The amount of CO2 production is obtained from a weight change of
the soda lime container plus the moisture collected in the second sulfuric acid container.
The whole body chamber offers fast response, accuracy and precision and
allows comparative measurements during sleep, at rest, before and after meals, and
during controlled periods of exercise. Ravussin and Rising (1992) reported that the
response time for their respiration chamber (a total volume of 19500 L) measured during
CO2 dilution was 3 min for 93% of the total response and 4 min for 99% of the total
response. In their review of techniques for the measurement of human energy expenditure,
Murgatroyd et al. (1993) reported that an accuracy of 1% in gas exchange measurement
by the whole body chamber is achievable. They further reported that an accuracy of 1.5
to 2% of energy expenditure was obtained if the precision of the gas exchange rate is
better than 4 mL/min over a period of 30 min in a chamber with a volume of 10000 L.
Although the whole body indirect calorimeters provide fast response, accuracy and
precision of measurement, satisfactory results are only obtained with careful chamber
design and construction, and by the choice of the best possible instrumentation.
Nevertheless, expensive cost and artificial environment (the subject is not under freeliving conditions) are still its disadvantages. Current design of the closed-circuit
volumetric replenishment indirect calorimeter employs an ultrasonic transducer to
monitor the volume level of a bellows that is used as an oxygen reservoir. As oxygen is
depleted, a microcomputer-controlled calibrated volumetric pulse of oxygen is delivered
to maintain the bellows at the preset reference level. The VDO 2 is calculated by
multiplication of the pulse volume by the number of pulses per minute. In addition, the
ultrasonic transducer is also used to monitor VT. Measurement of VDO 2 made by this
technique has been found to be quite accurate. A good correlation (r > 0.99) between
measured and actual VDO 2 has been reported by several authors (Branson et al., 1988;
Keppler et al., 1989). Advantages of the closed-circuit indirect calorimeters are that they
are not affected by fluctuations of FIO2 and by an FIO2 > 0.6. Details of technical
considerations have been discussed in Branson et al. (1995).
Doubly-labeled water method

The doubly labeled water (DLW) method was first introduced by Lifson, Gordon, and
McClintock in the 1950s (Lifson et al., 1955). They used the principles of indirect
calorimetry to measure total energy expenditure from the disappearance rates of two

10.3 Measurement of Body Heat: Calorimetry

23

stable isotopes: deuterium (2H) and oxygen 18 (18O). Since the early 1980s (Schoeller and
van Santen, 1982), the use of doubly labeled water has provided a technique where the
total energy expenditure can be measured in a free-living individual over 1 to 2 weeks. It
also has considerable appeal for field studies to estimate the energy cost of activities and
thermogenesis by subtracting the basal metabolic rate from the total energy expenditure.
The principle of the method is based on six assumptions listed as follows:
1. The body is in steady state; i.e. the volume of total body water pool remains
constant over time.
2. The turnover rates of water and carbon dioxide are constants over time.
3. The volume of labeled water distribution is only equal to total body water.
4. The isotopes are lost only as water and carbon dioxide.
5. Losses of labeled water and carbon dioxide in urine and exhaled carbon dioxide
have the same level of enrichment as in body water.
6. The background levels of the isotopes remain constant.

2H2O

H218O

deuterium water
oxygen-18 enriched water

18O

labeled water and

bicarbonate pools

2H

labeled water pool

2H

disappearance (k2) = rH2O

18O

disappearance (k18) =
rH2O + rCO2

k18 k2 = rCO2

Figure 10.13 Principle of the doubly labeled water method. r is the production rate, k represents
rate constants determined from the experiment.

Figure 10.13 shows the principle of the method. After a subject drinks a dose of
DLW, the deuterium mixes with the body water pool and the 18O mixes with both water

24

Body Temperature, Heat, Fat, and Movement

and bicarbonate pools. The production rate of CO2 is calculated as the difference
between 18O and 2H disappearance rates. The formula for this calculation is:
W
rCO 2 = VCO 2 = (k18 k 2 ) 22.4
2

(10.23)

where r (in L/h) is the production rate, W is the size of total body water in mol and k18
and k2 are fractional disappearance rates for 18O and 2H, respectively. They are
computed from measurements of the isotopic concentrations c18 and c2 using k = (1/c)
(c / t). The factor 2 in the denominator of Eq. (10.23) is the molar proportion of
oxygen in carbon dioxide in relation to water. In practice, Lifson et al. (1955) modified
Eq. (10.23) by including the fractional factor: f1 (0.93) for 2H2O gas/2H2O liquid, f2
(0.99) for H2 gas/H218O liquid, and f3 (1.04) for C18O2 gas/ H218O liquid to take a
natural fraction of the isotopes between different body fluids into account. Eq. (10.23)
becomes

or

W
f f D
rCO 2 = VDCO 2 =
(k18 k 2 ) 2 1 m 22.4
2 f 3 18
2 f 3
D
W
rCO 2 = VDCO 2 =
(k18 k 2 ) m 22.4
624
2.08

(10.24)

D is the rate of water loss as vapor in g/h. A minimum of two urine samples (one
where m
a few hours after the dose and another at the very end of the experiment) are collected to
compute the rate constants (k18 and k2) and the total body water (W). A number of crossvalidation studies have been performed by comparison with gas exchange analysis and
the whole body chamber under resting condition as well as during sustained heavy
exercise (see reviews in Ritz and Coward, 1995; Mclean and Tobin, 1987; Prentice et al.,
1991; Ravussin and Rising, 1992). Results showed that the doubly labeled water method
is a suitable technique for assessing energy expenditure in free-living circumstances with
an accuracy of 5%. However, large underestimates were found in obese subjects
(Ravussin and Rising, 1992). Advantages of the doubly labeled water method are that it is
noninvasive (oral dosing), nonintrusive (urine sampling), and suitable for noncompliant
subjects and field studies. In addition, deuterium distribution space can be used to
estimate total body water for an estimate of body composition (see section 10.4.1).
Nevertheless, the high cost for the expense of 18O-enriched water and technical
complexity are disadvantages of the doubly labeled water method. The amount of dosage
and its resulting cost depend on how precise the mass spectrometer is. In-depth
knowledge of the assumptions of the doubly labeled water technique is required to apply
it to various physiological and pathological situations.

10.5 Measurement of Body Movement

10.4

25

Measurement of Body Fat

Body fat measurement is performed during body composition studies. It is used in a wide
variety of fields, including human biology, medicine, sports science, and nutrition. Fat
measurements may be used to assess obesity in children and adults and to investigate the
effects of malnutrition. Body fat mass is defined as pure fattriglyceride fat in the body.
Adipose tissue consists of approximately 83% fat, 2% protein, and 15% water (Jensen,
1992). Fat free mass (FFM) is defined as lean body mass plus nonfat components of
adipose tissue. The assumption used by most body fat assessment methods is that the
body is made of two compartments, fat and fat free weight. The chemical composition of
the fat-free body is assumed to be relatively constant. The fat free body has a density of
1.1 g/cm3 at 37 C (Behnke et al., 1942) with water content of 72 to 74% (typical value
73.2%, Pace and Rathburn, 1945). Body fat, or stored triglyceride, is also assumed to be
constant with a density of 0.9 g/cm3 at 37 C (Mendez et al., 1960). All body fat methods
use either this twocomponent model or a further division of the fat and fat-free weight
into four compartments: water, protein, bone mineral, and fat (Brozek et al., 1963). The
various body fat measurement techniques may be further divided into two categories.
Some techniques are direct measures of body fat such as deuterium oxide dilution,
densitometry, and dual photon absorptiometry (DPA)/dual energy X-ray absorptiometry
(DEXA), whereas others are indirect and require crossvalidation with other methods.
Examples of indirect methods that will be discussed here are anthropometry and
bioelectrical impedance analysis (BIA). Although MRI and CT are able to directly
measure regional fat distribution, especially interabdominal fat content, high cost and
long scanning time still limit these two techniques to research studies.
10.4.1 Direct measurement of body fat

Direct measurement techniques of body fat such as densitometry (underwater weighing)

and deuterium oxide dilution are commonly used as reference methods for measuring
percent body fat. They are based on the two-compartment model and a constant hydration
factor of FFM, respectively. DPA/DEXA is another direct measuring technique that has
yet to be tested as a reference method. Underwater weighing provides accurate
measurement but it requires a well-trained technician, complicated equipment, and
laborious procedures. Consequently, it is not practical for field studies. In addition,
underwater weighing is not suitable for groups such as the handicapped, diseased people,
and young children. In contrast, the deuterium oxide dilution method is applicable for a
variety of groups. However, it also requires sophisticated equipment to measure the final
concentration of the isotope.
Deuterium oxide dilution

The use of the deuterium oxide dilution technique to measure body water to predict FFM
uses the fact that fat is mainly anhydrous and water is a relatively constant component

25

26

Body Temperature, Heat, Fat, and Movement

(73.2%) of FFM. It is based on the assumption that the isotope has the same distribution
volume as water, is exchanged by the body in a manner similar to water, and is nontoxic
in the amount used. During the experiment, subjects are asked to drink or are injected
with a known amount of the labeled water (deuterium oxide water). Then the amount of
the labeled water in saliva, urine, or plasma is determined after an equilibrium period (3
to 4 h). The calculation of total body water (TBW) volume is based on the simplified
relationship:
C1V1 = C2V2

(10.25)

where C1V1 is the amount of the isotope given, and C2 is the final concentration of the
isotope in saliva, urine, or plasma and V2 is the volume of TBW. Available techniques
for the assay of deuterium in aqueous solution are gas chromatography (see section
3.15.2), mass spectrometry (see section 3.12), and filter-fixed infrared absorptiometry.
The quantity of the isotope given depends on the analytical system used and the objective
of the research. Lukaski (1987) reported that a 10 g dose of deuterium oxide (99.7%
purity) mixed with ~300 mL of distilled-deionized water is used routinely in his
laboratory for the filter-fixed infrared absorptiometry. Some investigators have given an
oral dose of deuterium oxide of 1 g/kg body weight (Mendez et al., 1970) for gas
chromatography and an oral dose of 2 g D2O has been used for mass spectrometry
(Halliday and Miller, 1977).
For adult white individuals (Wellen et al., 1994), the calculation of fat free mass
from TBW volume is done by dividing V2 by 1.04 and 0.732 to account for the estimated
4% nonaqueous H+ exchange and water fraction of the FFM respectively.
V2
FFMTBW (kg ) =
(10.26)
(1.04 0.732 )
(weight FFMTBW )
%BFTBW =
100
weight

(10.27)

where FFMTBW is FFM derived from TBW (V2), weight is the subject's body weight,
and %BFTBW is the derived percent body fat from FFMTBW. For populations of low
(children) or high (obese) TBW, water fraction of the FFM (0.732) in Eq. (10.26) should
be adjusted to maintain an accurate estimation of %BFTBW. This change of TBW is
attributed to the fact that 15% of adipose tissue is water. In general, the accuracy
associated with measuring body fat by the TBW method is about 3% to 4% (Lukaski,
1987).
Densitometry (underwater weighing)

The most accurate method for estimating body fat is the direct measurement of body
density from underwater weighing. This method is often used as the reference method for
evaluating new indirect methods such as bioelectrical impedance analysis. Densitometry

10.4 Measurement of Body Fat

27

assumes that the human body has two distinct compartments: fat mass and fat free mass
(Siri, 1961; Brozek et al., 1963). The measurement of body density is accomplished by
measuring body volume (according to Archimedes' principle) and body weight (by the
difference between the subject's weight in air and weight in water). Archimedes' principle
states that the volume of a submerged object is equal to the amount of water the object
displaces. In a human subject, the apparent volume of the subject is determined by
correcting for the volume in the lung and the density of the water using the water
temperature at the time of the measurement.
The weighing device consists of a plastic chair suspended from a scale into a
huge tank containing water maintained at an average temperature of 35 C. The tare
weight for the chair and water temperature are measured before the experiment. During
the measurement, the subjects sit on the chair with minimal clothes and are submerged
under water. While remaining under water, the subjects exhale completely and hold their
breath. Underwater weight is recorded after bubbles stop coming from the subject's
mouth and oscillation of the scale due to the waves has subsided. Underwater weight is
recorded 10 times for each subject and the mean of the three highest weights is used.
Residual lung volumes are measured on land to account for its contribution to the
subject's buoyancy. The body density (BD) is expressed in g/cm3 and is calculated from
Eq. (10.28) (Guo et al., 1987):
Wa
BD =
(10.28)
Wa (W w T )
RV
Dw
where Wa = weight in air in grams, Ww = weight in water in grams, Dw = density of
water in g/cm3 at actual water temperature, RV = residual lung volume in cm3, and T =
tare weight in g. Density is then used to estimate the body fat with standard equations
(Siri, 1961; Brozek et al., 1963). The equation used by Siri to calculate the percent body
fat (%BF) is
%BF =

4.95
4.50
BD

(10.29)

%BF =

4.57
4.412
BD

(10.30)

and Brozek's equation is

Lukaski (1987) reported that Eqs. (10.29) and (10.30) give results within 1%
body fat if the body density is within 1.10 to 1.03 and that Siri's equation yields higher
values than the Brozeks equation for subjects with more than 30% body fat. Due to the
errors associated with variations of water and bone mineral contents, an error of 1 to 3%
in prediction of percent body fat with this method is expected (Lohman, 1990). In general,
the error is greater in populations with lower bone mineral content (children, women, and
old people) and populations with high bone mineral content (athletes). Noninvasive and

28

Body Temperature, Heat, Fat, and Movement

accurate measurement of body density, and good estimation of percent body fat have
made the underwater weighing a standard method for evaluation of other new and
portable techniques such as bioelectrical impedance analysis.
Dual photon absorptiometry (DPA) / dual energy X-ray absorptiometry (DEXA)

Another technique for direct measurement of body fat is dual photon absorptiometry or
dual energy X-ray absorptiometry. DPA/DEXA has been used in clinical settings to
diagnose osteoporosis and also to evaluate types of therapy for osteoporosis by directly
measuring bone mineral contents of the patients. Photon aborptiometry was first
introduced by Cameron and his colleagues (1962, 1963) to measure bone mineral content
(BMC) in the limbs with a single isotope (Iodine-125 at 27.3 keV or Americium-241 at
59.6 keV). This technique is called single photon absorptiometry (SPA) and is based on
the principle that bone mineral content is directly proportional to the amount of absorbed
photon energy by the target bone (see Figure 9.9). The necessity of constant thickness of
absorber and uncontrollable variations in transmission due to the presence of fat tissue
limit the measurement of BMC to bones only in the extremities. Mazess and his
colleagues (Mazess et al., 1970; Judy, 1970) developed dual photon absorptiometry to
measure contents of bone mineral and soft tissue in the appendicular parts of the skeleton
(e.g. femur) with the isotope Gadolinium-153 (153Gd, 44 and 100 keV). This method has
been applied to measure BMC in the axial skeleton (e.g. vertebrae) with the isotopes
Americium-241 (241Am, 59,6 keV) and Cesium-137 (137Cs, 662keV) by Roos and
Skoldborn (1974). The underlying principle of DPA/DEXA for the measurement of body
fat is that the ratio of absorbance of the two different energy level photons (R = L/H) is
linearly related to the percent of fat in the soft tissue of the body. The mass attenuation
coefficients for bone mineral and soft tissues are constant and are not affected by the
amount of bone and soft tissues along the path of the photon beam. They depend on the
energy level of the beam. The basic equations for DPA are as follows (Sorenson et al.,
1989):
N
ln 0 L
NL

= sL M s + bL M b,

(10.31)

N
ln 0 H
NH

= sH M s + bH M b,

(10.32)

where N0L and NL are the number of incident and transmitted low energy photons, N0H
and NH are the number of incident and transmitted high energy photons, sL and sH are
the mass attenuation coefficients of soft tissue at the low and high energies, bL and bH
are the same quantities for bone mineral, and Ms and Mb are the area densities in g/cm2
of soft tissue and bone mineral in the beam path. By solving Eqs. (10.31) and (10.32)
simultaneously, Mb can be derived as:

10.4 Measurement of Body Fat

N 0H
N
ln 0 L
R ln
NH
NL

Mb =
(R bH bL )

29

(10.33)

where R = sL/sH. Based on the same basic equations, DEXA uses a constant voltage
X-ray generator and k-edge filters (made of cerium) to separate the X-ray beam into two
energy levels, or a switching-pulse system that rapidly alternates the constant voltage Xray generator to produce two beams with high and low energy simultaneously. Current
examples of the former type of densitometer include Norland XR-36 (Norland Medical
System, Inc., Fort Atkinson, WI) and Lunar DPX series (Lunar Radiation Corp., Madison,
WI). An example of the latter type of densitonometer is the Hologic QDR-1000 series
(Hologic, Inc., Waltham, MA). In recent years, DEXA has become the main stream in the
densitometer industry. DEXA scans faster and has better image quality over DPA.
Most of the current commercially available densitometers usually include a
scanning and examination table, X-ray system (scanner and scintillation detector), a
computer system with additional software to control and calculate BMC and body
composition, and positioning or calibration accessories. During the measurement, the
patient needs to lie supine on the table and a series of transverse scans are made from
head to toe at 1 cm intervals (Mazess et al., 1990). Measurements take 10 to 20 min for a
total body measurement or 1 to 5 min for a regional scan. Typical transverse scan image
area is around 60 cm 200 cm. The value of R (weighted for mass of tissue in each pixel)
is determined for the number of pixels containing the soft tissue alone. The %BF can be
derived from the R value through application of a regression equation of %BF and the R
value resulting from the calibration phantom provided by the manufacturer. Total body
bone mineral content is calculated from the bone-containing pixels. These calculations
are usually processed by the software provided by the manufacturer. Technology of
automatic exposure control is also applied to DEXA to assure that each pixel receives the
proper exposure regardless of tissue thickness. In general, the precision error for the
percent body fat is within approximately 2% (Lukaski, 1987; Mazess et al., 1990; Jensen,
1992; Prentice, 1995). The advantages of DPA/DEXA include portability, capability of
measuring total body or regional composition of bone mineral and body fat, high
reproducibility (3 to 5%), and less dependence on the technician's skills and experience.
Narrow width table and size limitation prohibit using the DPA/DEXA technique to assess
very obese subjects. Radiation exposure of DPA/DEXA is between 2 and 10 mrem.
10.4.2 Indirect measurement of body fat

Indirect measurement techniques of body fat such as anthropometry and BIA require
empirically derived regression equations to estimate body fat or total body water (TBW).
These techniques are suitable for population studies where individual distinction is less
critical. They often need to be validated by reference methods such as densitometry and
the isotope dilution method. Anthropometry is a rapid and inexpensive way to evaluate
nutritional status for a population in a field study, but it requires a skilled technician and

30

Body Temperature, Heat, Fat, and Movement

anthropometrist to achieve accurate measurement. Single or multiple frequency BIA also

provides a rapid and simple method to allow the investigator to predict TBW and
extracellular water (ECW). In general, both techniques are less likely to be useful for
prediction of change in body composition.
Anthropometry (Skinfold thickness)

Anthropometry is defined as measurement of body size. It includes measurement of body

circumferences, skinfolds, weight, stature, etc. Measurement of skinfolds is one of the
most commonly used anthropometric techniques. It is based on two assumptions: the
subcutaneous adipose tissue represents a fairly constant proportion of the total body fat
and the selected measurement sites represent the average thickness of subcutaneous
adipose tissue (Lukaski, 1987). Skinfold thickness is the thickness of a double layer of
skin and subcutaneous adipose tissue that is pulled away from the muscle at selected sites.
Skinfold thickness can be measured using special devices such as the Lange caliper
(Lange and Brozek, 1961, see Figure 10.14(a)) and the Holtain (the Harpenden) caliper
(Tanner and Whitehouse, 1955). The main uses for skinfold thickness are to estimate
body fat (percent body fat, or %BF) and the anatomical distribution of fat tissue for
population surveys. Common sites for measurement of skinfold thickness are the chest,
midaxilla, abdomen, supraillium, triceps, subscapular, and thigh.
Shoulder

Triceps
skinfold

Lange
caliper
a

Figure 10.14 (a) Lange skinfold caliper used for assessing thickness of subcutaneous fat. (b)
Illustration of an example of skinfold measurement, triceps skinfold taken on the midline posterior
surface of the arm over the triceps muscle.

During the measurement of skinfold thickness, a double layer of skin and

subcutaneous fat tissue is grasped between the thumb and forefinger and pulled forward
from the site by 1 cm to allow the jaw of the caliper, maintaining a constant jaw pressure
of 10 g/mm2, to impinge on the skin. A distance of at least 1 cm should be maintained
between the fingers and the caliper arms to prevent any interference from the pressure of
the fingers. The calipers should be applied perpendicular to the body surface and

10.4 Measurement of Body Fat

31

perpendicular to the longitudinal axis of the skinfold (see Fig. 10.14(b)). The release of
the pressure of the caliper arms requires the measurement to be made within 4 s to
standardize differences among subjects in compressibility of skinfolds. The sequence of
measurements is repeated three times to reduce intrameasurer bias. For a more complete
review of specific anthropometric measurement techniques, see the Anthropometric
Standardization Reference Manual (Lohman et al., 1988).
Many regression equations are available for the prediction of body density and
thus body fat from skinfold thickness measurements. These equations are generally valid
for adult Caucasian populations, but overestimate body fat in elderly individuals. For
example, Jackson and Pollack (1978) and Jackson et al. (1980) published equations for
estimating body density in adult men, Eq. (10.34), and women, Eq. (10.35), aged 20 to 50
years.
BD = 1.1093800 0.0008267 ( X 1) + 0.0000016 ( X 1)2 0.0002574 ( X 3 )

(10.34)

BD = 1.0994921 0.0009929 ( X 2 ) + 0.0000023( X 2 ) 0.0001392 ( X 3)

(10.35)

where BD is body density, X1 is the sum of chest, abdomen, and thigh skinfolds (mm),
X2 is the sum of triceps, thigh, and suprailium skinfolds in mm, and X3 is the age.
Correlation coefficients (r) for Eqs. (10.34) and (10.35) are 0.905 and 0.842, respectively.
These body densities can subsequently be used for percent body fat estimation with Eq.
(10.27).
The advantages of body fat prediction from measurement of skinfold thickness
are that it is simple, inexpensive, and noninvasive. However, the accuracy and reliability
of skinfold thickness measurement depends on the skill of the measurer, interaction
between the measurer and the subject, and the characteristic of the subject. The potential
error associated with the use of skinfolds, assuming the correct equation is used, the
measurer is well trained, and correct skinfold sites chosen, is 3 to 4% (Lohman, 1990). In
addition, body fat of edematous (excess fluid in the tissues) patients is often
overestimated and very obese subjects cannot fit in the standard calipers.
Bioelectric impedance analysis

Bioelectrical impedance analysis (BIA) for body fat measurement is used for
measurement of fat free mass of the body. It is also based on the two-compartment model
(in parallel) as mentioned earlier. Fat free mass is comprised mainly of electrolytecontaining water (73.2%) in the body; it represents the main path of conduction for the
applied current. In contrast, fat mass behaves as an insulator. Therefore, the impedance of
the body is controlled by fat free mass. The method of BIA is based on the principle that
the impedance of an isotropic conductor is related to the length and cross-sectional area
of the conductor for a constant signal frequency. Prediction equations are often generated
using BIA values to estimate fat free mass. With a constant signal frequency and a
relatively constant conductor configuration, the impedance to an ac current can be related
to the volume of the conductor:

32

Body Temperature, Heat, Fat, and Movement

Z=

L2
V

(10.36)

where Z is the impedance in ohms (), is the resistivity in cm, L is the conductor
length in cm, and V is the volume in cm3. A general procedure to measure the whole
body impedance is to let the subject, 2 h after eating and 30 min after voiding, lie supine
on a table with arms and legs sufficiently apart. Current injector electrodes are placed in
the middle of the dorsal surfaces of the right metacarpalphalangeal (hand) and
metatarsalphalangeal (foot) joints, respectively. Detector electrodes are positioned in the
midline of the dorsal surfaces of right wrist and ankle. Measurements of body resistance
and reactance are made on ipsilateral and contralateral sides of the body (as shown in
Figure 10.15). The lowest value of resistance for each subject is used to predict fat free
mass (FFM). Hoffer et al. (1969) first used four surface electrodes (tetrapolar) for the
study of TBW. The tetrapolar technique is used to minimize contact impedance and skin
electrode interaction. They introduced 100 A of ac current at 100 kHz in 20 healthy
subjects and 34 patients whose TBW were measured with the isotope dilution method
(see section 10.4.1) and showed that (stature)2/Z was the best predictor of TBW. The
correlation coefficients (r) of (stature)2/Z against TBW for 20 healthy subjects and 34
patients were 0.92 and 0.93, respectively. Lukaski et al. (1985) demonstrated in his total
body impedance study in 37 men that the resistance, R, is a better predictor than the
impedance (square root of R2 + Xc2) when the reactance, Xc, is small. They used a
current source of 800 A at 50 kHz (RJL Systems, Detroit, MI). Consequently, the best
predictor of TBW (r = 0.95) and FFM (r = 0.98) was (stature)2/R. Kushner and Schoeller
(1986) cross-validated the prediction of TBW with BIA by generating a prediction
equation from a sample of 40 nonobese adults and applying it to 18 obese patients. They
found that (stature)2/R was a good predictor of TBW (r = 0.96). Segal et al. (1985) also
confirmed the significant role of (stature)2/Z as a predictor of FFM (r = 0.912) with 75
male and female subjects ranging from 4.9 to 54.9% body fat, but reported
overestimation of FFM for obese subjects. Because fat free mass includes virtually all
water in the body, the formula that is used to predict TBW via BIA is often used to
estimate FFM. A general model that is often used to predict FFM or TBW in kg is:

Stature 2
FFM or TBW = C1

+ C 2 (Weight ) + C 3

(10.37)

where C1 and C2 are regression coefficients of the independent variables Stature2/R in

cm2/ and Weight in kg, respectively. C3 is the intercept of this regression equation.
These constants vary with different sample groups. The equation

FFM

% BF = 100
Weight

(10.38)

10.4 Measurement of Body Fat

33

has been used to estimate percent body fat. Segal et al. (1988) and Gray et al. (1989) have
suggested that prediction of FFM can be enhanced by the use of gender and fatness
specific equations.
Excitation ac current source (I)
Detected voltage drop (E)

Figure 10.15 In resistance measurement on the ipsilateral side of the body current flows through
one arm, the trunk, and one leg.

Current commercially available BIA instruments (RJL Systems, Detroit, MI;

Dieto-System, Milan, Italy; etc.) are often equipped with multiple frequency function and
specialized software. They provide empirically derived regression equations and are
capable of estimating TBW, FFM, %BF, and extracellular water (ECW). Theoretically
TBW is modeled as a parallel RC circuit with the cell membrane as a capacitor to
separate intracellular and extracellular fluids. At low frequency (~1 to 5 kHz), the current
only reaches the extracellular fluid while at high frequency (~50 kHz and higher) it
penetrates both fluids. Therefore, at low frequency the impedance can be used to estimate
ECW. Deurenburg (1995) reported in his study of 103 healthy males and females, aged
1951 years that ECW was best predicted by resistance measured at 1 kHz (r = 0.93) and
TBW was best estimated by resistance measured at 100 kHz (r = 0.95). Regression
coefficients (C1, C2, and C3) estimated from his study for TBW and ECW were 0.368,
0.133, 2.1 and 0.177, 0.073, 2.0, respectively. Segal et al. (1991) also confirmed the
significant role of resistance measured at 5 kHz and 100 kHz for prediction of ECW (r =
0.93) and TBW (r = 0.95) in 36 healthy males. Regression coefficients (C1, C2, and C3)
estimated from their study for TBW and ECW were 0.455, 0.14, 3.43 and 0.284, 0.112,
6.115, respectively. They cross-validated on two randomly selected subsets (18 each) and
showed that the prediction equations were reproducible and valid.
BIA offers the advantage of rapid, safe, noninvasive measurements of percent
body fat. In addition, multifrequency BIA can differentiate extracellular water (ECW)
and total body water (TBW). Problems with the BIA method include that there is no
generally accepted regression equation as a reference for all sample groups and BIA
instruments and that accurate prediction of a change of ECW is quite difficult.

10.5 Measurement of Body Movement

Body movement is studied to learn how to analyze the movement of the body and to
understand its underlying principles. It is useful for improving performance of motor

34

Body Temperature, Heat, Fat, and Movement

skills (such as skating) and providing efficient methods of using the body in daily life
skills (e.g., walking and running). Analysis of body movement requires observation and
measurement. Many methods are available, ranging from simple visual observation to
complicated computerized systems. Typical research or clinical gait analysis laboratory
setups include goniometers, accelerometers, electromyography (EMG, see Chapter 7),
force plates, and kinematic (motion) analysis systems. A kinematic analysis system varies
from simple and subjective visual observation to more expensive and complicated video
or optoelectronic systems that can provide objective, quantitative data of threedimensional motion of selected points on the body. With additional force plate outputs,
the kinematic analysis system can be used as a kinetic analysis system to provide force
and moment information for human limb movement. In this section, we discuss working
principles and applications for goniometers, accelerometers, and video and optoelectronic
kinematic analysis systems. For additional information, refer to texts by Allard et al.
(1995), Whittle (1996), Winter (1990), and Medved (2001).
10.5.1 Goniometers and accelerometers
Goniometers

A goniometer is a device that measures relative angle of a joint that connects two body
segments. The goniometer (see Figure 10.16(a)) consists of two attachment arms, which
are fixed to two limb segments of the subject on either side of a joint to measure the joint
angle without interfering with natural movement. Traditionally, a goniometer uses a
resistance potentiometer to convert changes in rotation to an electric output proportional
to the positions of two attachment arms. The potentiometer is aligned with the joint axis
to establish the zero position and is calibrated to measure the joint angle in degrees.
Multiaxial measurements of the joint are possible with two or three potentiometers
mounted orthogonally in different planes. The output of the goniometer provides
continuous analog data of joint motion at relatively low cost. However, to ensure accurate
measurements, fixation to the body with cuffs around the soft tissues must be performed
carefully to prevent off-axis problems (Chao, 1980). The measurement error caused by
the translation of the joint center during movement can be improved by replacing the
rigid arms of the traditional goniometer with a flexible parallelogram linkage designed to
accommodate the motion change outside the measurement plane (Thomas and Long,
1964). A triaxial parallelogram electrogoniometer (see Figure 10.16(b)) has been used to
record motions in more than one plane for the lower limb (Isacson et al., 1986). Strain
gages have also been used as an alternative to the potentiometer. They deform when they
are stretched or compressed and the electric output is proportional to the change in joint
rotation. The working principle of strain gage and its related circuitry is discussed in
section 8.3.1.

10.5 Measurement of Body Movement

Plastic ribbons
Connecting rods

Vi
Vout = k

35

Parallelogram
linkage

Knee joint
Triaxial
goniometer

a
b
Figure 10.16 (a) A goniometer attached to the shank and thigh to measure knee rotation. Vi is the
input voltage. Vout is the output voltage that is proportional to the angle of knee rotation. (b)
Subject wearing a triaxial goniometer on knee joint.

Accelerometers

Accelerometers are instruments that measure acceleration. Basically, they all contain a
small mass and make use of Newton's law (F = ma) to measure the force that is required
to accelerate the known mass, as shown in Figure 10.17. There are two classes of
accelerometers used in the analysis of body movement: strain gage (usually piezoresistive)
and piezoelectric accelerometers. A strain gage accelerometer consists of strain gages
(wires) bonded to a cantilevered mass and a base to which the cantilever beam is attached.
When the base is accelerated, the cantilever beam deforms due to the inertia of the mass,
thus changing the strain in the wires. The strain in the wires changes their resistances. We
measure the change of resistance with a Wheatstone bridge circuit (see section 9.11.2),
which requires a differential amplifier. The resulting electric output is proportional to the
acceleration of the mass. A piezoresistive accelerometer works under the same principle
as a strain gage accelerometer. Instead of using strain sensitive wire, it uses piezoresistive
strain elements as sensors. Piezoelectric accelerometers measure the force directly. The
piezoelectric sensor converts the force produced by the acceleration to an electric charge
or voltage (see Eq. (8.7) and section 8.5.2 for the working principle and its charge
amplifier). For the analysis of body movement, the accelerometer is attached to a body
segment at a specific point to measure acceleration in one direction. Two- or threedimensional measurements are possible with several accelerometers grouped together
orthogonally. The use of accelerometers for the analysis of body movement provides an
alternative way to measure the velocity and displacement of the limb segment when the
initial values of the limbs velocity and displacement are known (Morris, 1973). Some
practical limitations prevent widespread use of the accelerometer in body motion analysis:
necessity to exclude the effect of the field of gravity from true kinematic acceleration,
difficulty in extracting the rotational acceleration, and low frequency noise caused by the
baseline drift in the measurement output.

36

Body Temperature, Heat, Fat, and Movement

Strain

Seismic
mass

Cantilever

Figure 10.17 Vertical acceleration of the accelerometer frame bends the cantilever beam because
the seismic mass remains at rest. Voltage output (V) is proportional to the acceleration (a). E is the
supply voltage.

10.5.2 Video and optoelectronic systems

Gait can be measured on a single film using sequential strobe lights and a camera. Gait
can be measured by video recorder. Figure 10.18 shows that a video or optoelectronic
kinematic analysis system consists of active or passive markers placed at selected skin
surface locations on bony landmarks, sensors that track markers' positions, and a
computer system to control the timing for active marker display and signal processing of
markers' tracings. The sensors, either optoelectronic circuitry or a video system, collect
kinematic data from active (infrared LED) or passive markers and low-pass filter the data
before an analog-to-digital converter (ADC), which feeds to a computer for later
computation of the position and velocity of the rigid body in three-dimensional (3-D)
space. A computerized analysis system often yields graphic displays that can provide
information on joint positions, joint angles, segment angles, velocity, and acceleration as
a function of time.
Camera 4

Camera 1

Markers
EMG
transmitter
(option)

Force
platforms
(option)

Camera
control
and
Interface

Walk way

Computer

Display
system

Analog interface
Camera 3

Camera 2

A/D
converter

EMG
receiver
Figure 10.18 An example of a gait analysis setup includes a four-camera kinematic system, two
force platforms, and an electromyogram (EMG) telemetry system.

10.5 Measurement of Body Movement

37

There are many commercially available video or optoelectronic kinematic

analysis systems (Walton, 1990). They are different in the way that they combine video
or optoelectronic systems with passive or active markers. For examples, Vicon (Oxford
Metrics Ltd., Botley, Oxford, UK) and ELITE Plus (Bioengineering Technology &
Systems, Milan, Italy) are video systems with passive (reflective) markers, and Selspot II
(Selcom Inc., Southfield, MI), OPTOTRACK (Northern Digital Inc., Waterloo, Ontario,
Canada), and CODA (Charnwood Dynamics Ltd., Leics, UK) are optoelectronic systems
with active (infrared emitting diode) markers. All current kinematic analysis systems are
precalibrated and have full 3-D motion tracking and display features in real time. In order
to achieve real-time tracking of markers in 3-D, problems of real-time 2-D tracking with
marker identity maintained and 2-D-to-3-D transformations in real time have to be solved.
The speed and convenience of tracking differs from one system to another, and also
between software options from the same manufacturer. They all depend on how many
video cameras and markers they use and how often they sample the data. Passive markers
that allow automatic processes are recognized by contrast (Vicon), shape (ELITE Plus),
and color (VIDEOMEX, COLUMBUS INSTRUMENTS, Columbus, OH) features
during data analysis. They are suitable for sports studies or whenever large and fast
movements are to be recorded because the subject is not hampered by attached wires and
battery packs carried on belts as active markers. Active markers are infrared lightemitting diodes (IRLED, 800 nm) that allow simple labeling and high frequency
sampling. The IRLEDs are often arranged in sequence and pulsed at different times (only
one at a time) by a control unit. Most of the video systems use either stroboscopic
illumination or CCD cameras (see section 6.2.3 for more information) or both to avoid
the smearing when the markers are moving. They are often synchronized by the camera
control and interface unit to collect data with analog devices such as force plates, and
EMG for movement kinetics.
Optoelectronic systems are similar, in principle, to video systems. They use
photodetectors to focus the pulsed infrared signals from body markers from associated
body landmarks. The determination of 3-D spatial coordinates from several 2-D sets of
information is often accomplished by direct linear transformation (DLT) method (Abdel
Aziz and Karara, 1971). For m markers and n cameras, the DLT method determines the
3-D spatial coordinates of a marker from the 2-D coordinates of the marker, i (i = 1, ,
m) on camera j, (j = 1, , n) by
x ij =
y ij =

A j xi + B j yi + C j z i + D j
I j xi + J j yi + K j z i + 1
E j xi + F j yi + G j z i + H j
I j xi + J j yi + K j z i + 1

where for a given marker i:

xij = x coordinate of marker i on camera j,
yij = y coordinate of marker i on camera j,
xi = x coordinate of marker i in 3-D space,
yi = y coordinate of marker i in 3-D space,

(10.39)
(10.40)

38

Body Temperature, Heat, Fat, and Movement

zi = z coordinate of marker i in 3-D space,

Aj to Kj = coefficients in the transformation of marker i.
Figure 10.18 shows an example of a gait analysis laboratory, including a four
camera kinematic system, two force platforms, and an electromyogram (EMG) telemetry
system. For more information, refer to texts in Allard et al. (1995) and Walton (1990).

10.6 References
Allard, P., Stokes, I. A., and Blanchi, J.-P. 1995. Three-Dimensional Analysis of Human
Movement. Champaign IL: Human Kinetics.
Abdel-Aziz, Y. I. and Karara, H. M. 1971. Direct linear transformation from comparator
coordinates into object space coordinates in close-range photogrammetry. Proc.
Close-range Photogrammetry. Am. Soc. of Photogrammetry. 118.
Bair, D. N. and Davies, A. 1995. Temperature probes and principles of thermometry. In
R. L. Levine and R. E. Fromm, Jr. (eds.) Critical Care Monitoring: From PreHospital to the ICU. St. Louis: Mosby.
Behnke, A. R., Feen, B. G., and Welham, W. C. 1942. Specific gravity of healthy men.
JAMA, 118: 49598.
Benzinger, T. H. and Kitzinger, C. 1949. Direct calorimetry by means of the gradient
principle. Rev. Sci. Instrum., 20: 84960.
Branson, R. D., Lacy, J., and Berry, S. 1995. Indirect calorimetry and nutritional
monitoring. In R. L. Levine and R. E. Fromm, Jr. (eds.) Critical Care Monitoring:
From Pre-Hospital to the ICU. St. Louis: Mosby.
Branson, R. D., Hurst, J. M., Davis, K. Jr., and Pulsfort, R. 1988. A laboratory evaluation
of the Biergy VVR calorimeter. Resp. Care, 33: 3417.
Brozek, J., Grande, I., Anderson, J. T., and Keys, A. 1963. Densitometric analysis of
body composition: revision of some quantitative assumption. Ann. NY Acad. Sci., 40:
1103.
Cameron, J. R., Grant, R., and MacGregor, R. 1962. An improved technique for the
measurement of bone mineral content in vivo. Radiology, 78: 117.
Cameron, J. R. and Sorenson, J. 1963. Measurement of bone mineral in vivo: an
improved method. Science, 142: 2302.
Chao, E. Y. 1980. Justification of triaxial goniometer for the measurement of joint
rotation. J. Biomechanics, 13: 9891006.
Cobbold, R. S. 1974. Transducers for Biomedical Measurements: Principles and
Applications. New York: John Wiley & Sons.
Deurenburg, P. 1995. Multi-frequency impedance as a measure of body water
compartments. In P. S. W. Davies and T. J. Cole (eds.) Body Composition
Techniques in Health and Disease. Cambridge: Cambridge University Press.
Fraden, J. 1991. Noncontact temperature measurements in medicine. In D. L. Wise (ed.)
Bioinstrumentation and Biosensors. New York: Marcel Dekker.
Gray, D. S., Bray, G. A., Gemayel, N., and Kaplan, K. 1989. Effect of obesity on
bioelectrical impedance. Am. J. Clin. Nutr., 50: 25560.

10.6 References

39

Guyton, A. C. and Hall, J. E. 1996 Textbook of Medical Physiology. 9th ed. Philadelphia:
W. B. Saunders.
Halliday, D. and Miller, A. G. 1977. Precise measurement of total body water using trace
quantities of deuterium oxide. Biomed. Mass Spectrom, 4: 827.
Hoffer, E. C., Meador, C. K., and Simpson, D. C. 1969. Correlation of whole body
impedance with total body water. J. Appl. Physiol., 27: 5314.
Isacson, J., Cransberg, L., and Knuttsson, Z. 1986. Three-dimensional electrogoniometric
gait recording. J. Biomech., 19: 62735.
Jackson, A. S. and Pollock, M. L. 1978. Generalized equations for predicting body
density of men. Br. J. Nutr., 40: 497504.
Jackson, A. S., Pollock, M. L., and Ward, A. 1980. Generalized equations for predicting
body density of women. Med. Sci. Sports Exerc., 12: 17582.
Jensen, M. D. 1992. Research techniques for body composition assessment. J. Am.
Dietetic Assoc., 92: 45460.
Judy, P. E. 1970. A Dicromatics Attenuation Technique for the in vivo Determination of
Bone Mineral Content. Ph.D. Thesis, University of Wisconsin, Madison, WI.
Keppler, T., Dechert R., Arnoldi, D. K., Filius R., and Bartlett, R. H. 1989. Evaluations
of Waters MRM-6000 and Biergy VVR closed-circuit indirect calorimeters. Resp.
Care, 34: 2835.
Kushner, R. F. and Schoeller, D. A. 1986. Estimation of total body water by bioelectrical
impedance analysis. Am. J. Clin. Nutr., 44: 41724.
Lange, K. O. and Brozek, J. 1961. A new model of skinfold caliper. Am. J. Phys.
Anthropol., 19: 989.
Lifson, N., Gordon, G. B., and McClintock, R. 1955. Measurement of total carbon
dioxide production by means of D218O. J. Appl. Physiol., 7: 70410.
Lohman, T. G. 1990. Body composition assessment in sports medicine. Sports Med.
Digest., 12: 12.
Lohman, T. G., Roche, A. F., and Martorell, R. (eds.) 1988. Anthropometric
Standardization Reference Manual. Champaign, IL: Human Kinetics Books.
Lukaski, H. C., Johnson, P. E., Bolonchuk, W. W., and Lykken, G. I. 1985. Assessment
of fat-free mass using bioelectrical impedance measurements of the human body. Am.
J. Clin. Nutr., 41: 41017.
Lukaski, H. C. 1987. Methods for the assessment of human body composition: traditional
and new. Am. J. Clin. Nutr., 46: 53756.
Levine, R. L. and Fromm, R. E., Jr. (eds.) 1995. Critical Care Monitoring: From PreHospital to the ICU. St. Louis: Mosby.
Mazess, R. B., Ort, M., Judy, P., and Mather, W. 1970. Absorptiometric bone mineral
determination using 153Gd. Proc. CONF-700515, Bone Measurement Conf.,
Chicago, IL, 308.
Mazess, R. B., Barden, H. S., Bisek, J. P., and Hanson, J. 1990. Dual-energy X-ray
absorptiometry for total-body and regional bone-mineral and soft-tissue composition.
Am. J. Clin. Nutr., 51: 110612.
McLean, J. A. and Tobin, G. 1987. Animal and Human Calorimetry. Cambridge, UK:
Cambridge University Press.
Medved V. 2001. Measurement of Human Locomotion. Boca Raton, FL: CRC Press.

40

Body Temperature, Heat, Fat, and Movement

Mendez, J., Keys, A., Anderson, J. T., Grande, F. 1960. Density of fat and bone mineral
of mammalian body. Metabolism, 9: 4727.
Mendez, J., Procop, E., Picon-Reategui, F., Akers, R., and Buskirk, E. R. 1970. Total
body water by D2O dilution using saliva samples and gas chromatography. J. Appl.
Physiol., 28: 3547.
Milsum, J. H. 1966. Biological Control System Analysis. New York: McGraw-Hill.
Morris, J. R. W. 1973. Accelerometrya technique for the measurement of human body
movements. J. Biomechanics, 6: 72936.
Murgatroyd, P. R., Shetty, P. S., and Prentice, A. M. 1993. Techniques for the
measurement of human energy expenditure: a practical guide. Int. J. Obesity, 17:
54968.
Pace, N. and Rathburn, E. N. 1945. Studies on body composition. III The body water and
chemically combined nitrogen content in relation to fat content. J. Biol. Chem., 158:
68591.
Prentice, A. M. 1995. Application of dual-energy X-ray absorptiometry and related
techniques to the assessment of bone and body composition. In P. S. W. Davies and
T. J. Cole (eds.) Body Composition Techniques in Health and Disease. Cambridge
UK: Cambridge University Press.
Prentice, A. M., Diaz, E. O., Murgatroyd, P. R., Goldberg, G. R., Sonko, B. J., Black, A.
E., and Coward, W. A. 1991. Doubly labeled water measurements and calorimetry in
practice. In R. G. Whitehead and A. Prentice (eds.) New Techniques in Nutritional
Research. New York: Academic Press.
Ravussin, E. and Rising, R. 1992. Daily energy expenditure in humans: measurements in
a respiratory chamber and by doubly labeled water. In J. M. Kinney and H. N.
Tucker (eds.) Energy Metabolism: Tissue Determinants and Cellular Corollaries.
New York: Raven Press.
Ritz, P. and Coward, W. A. 1995. Doubly labeled water measurement of total energy
expenditure. Diabetes & Metabolism, 21: 24151.
Roos, B. O. and Skoldborn, H. 1974. Dual photon absorptiometry in lumbar vertebrae. I.
theory and method. Acta Radiologica: Therapeutics, 13: 26680.
Schoeller, D. A. and van Santen, E. 1982. Measurement of energy expenditure in humans
by doubly-labeled water method. J. Appl. Physiol., 53: 9559.
Segal, K. R., Gultin, B., Presta, E., Wang, J., and van Itallie, T. B. 1985. Estimation of
human body composition by electrical impedance methods: a comparative study. J.
Appl. Physiol., 58: 156571.
Segal, K. R., Van Loan, M., Fitzgerald, P. I., Hodgdon, J. A., Van Itallie, T. B. 1988.
Lean body mass estimated by bioelectrical impedance analysis: a four-site crossvalidation study. Am. J. Clin. Nutr., 47: 714.
Segal, K. R., Burastero, S., Chun, A., Coronel, P., Pierson, R. N., and Wang, J. 1991.
Estimation of extracellular and total body water by multiple-frequency bioelectricalimpedance measurement. Am. J. Clin. Nutr., 54: 269.
Siri, W. E. 1961. Body composition from fluid spaces and spaces, analysis, of methods.
In J. Brozek and A. Henschel (eds.) Measuring Body Composition. Washington DC:
National Academy Sciences, 22344.
Tanner, J. M. and Whitehouse, R. H. 1955. The Harpenden skinfold caliper. Am. J. Phys.
Anthropol., 13: 7436.

10.6 References

41

Thomas, D. H. and Long, C. 1964. An electrogoniometer for the fingera kinesiologic

tracking device. Am. J. Med. Elect., 3: 96100.
Walton, J. S. (ed.) 1990. Mini-Symposium on Image-Based Motion Measurement. SPIE1356, Bellingham, WA: The Society of Photo-Optical Instrumentation Engineers.
Webb, P. 1985. Human Calorimeters. New York: Praeger.
Webster, J. G. (ed.) 1998. Medical Instrumentation: Application and Design. 3rd ed. New
York: John Wiley & Sons.
Weir, J. B. de V. 1949. New methods for calculating metabolic rate with special
reference to protein metabolisms. J. Physiol., 109: 19.
Weissman, C., Sardar, M. S., and Kemper, M. 1990. In vitro evaluation of a compact
metabolic measurement instrument. JPEN, 14: 21621.
Wellens, R., Chumlea, W. C., Guo, S., Roche, A. F., Reo, N. V., and Siervogel, R. M.
1994. Body composition in white adults by dual-energy X-ray absorptiometry,
densitometry, and total body water. Am. J. Clin. Nutr., 54: 54755.
Whittle, M. W. 1996. Gait Analysis: An Introduction. 2nd ed. Oxford: ButterworthHeinemann.
Winter, D. A. 1990. Biomechanics and Motor Control of Human Movement. 2nd ed. New
York: John Wiley & Sons.

10.7 Problems
10.1
10.2
10.3
10.4
10.5
10.6
10.7
10.8

10.9
10.10
10.11

Calculate the value of the thermistor temperature coefficient for T = 298 K

and = 3000 K.
Design a circuit that converts the resistance change of a thermistor to a voltage,
and then amplifies it to measure temperature.
Explain the Seebeck effect of a thermocouple and why and how to use it for
temperature measurement.
Design a circuit that uses an electronic cold junction and thermocouple, then
amplifies it to measure temperature.
Explain the operation of a pyroelectric infrared thermometer, sketch its block
diagram, and sketch the response after the shutter opens.
Compare direct and indirect calorimetry.
Explain how to use the gradient layer calorimeter for heat measurement.
Figure 10.6 shows a simple gradient layer calorimeter. Sketch an expanded
gradient layer calorimeter that includes ventilation. Show location of all sensors
and explain how they work. Give a symbol for the output of each sensor.
Explain by means of equations, block diagrams and/or words how to obtain the
quantities given in equations (10.7) through (10.11) from your sensor outputs.
The oxygen consumption is 0.25 L/min. The carbon dioxide production is 0.06
L/min. Calculate the energy expenditure. Give units.
Define %BF and explain its purpose.
Explain how to measure %BF using densitometry.

42
10.12
10.13
10.14
10.15
10.16

Body Temperature, Heat, Fat, and Movement

Describe how to use the bioelectrical impedance method for body fat
measurement.
Draw the circuit for a resistance potentiometer goniometer with an amplifier
with a gain of 2
Draw the circuit for a 4-arm piezoresistive strain gage accelerometer and
amplifier.
Draw the circuit for a piezoelectric accelerometer and amplifier to pass
frequencies higher than 1 Hz.
You want to analyze human locomotion; list the devices you need and explain
why.