Proceedings of IEEE International Conference on Innovations in Electrical, Electronics, Instrumentation and Media Technology

ICIEEIMT 17

Performance Measures on Blood Pressure and

Heart Rate Measurement from PPG Signal for
Biomedical Applications
Melvin Johnson. A R. Jegan X. Anitha Mary
PG Scholar, Embedded Systems Assistant Professor Assistant Professor
Department of Electrical Technology, Department of Electrical Technology, Department of Electrical Technology,
Karunya University, Coimbatore, Karunya University, Coimbatore, Karunya University, Coimbatore,
India India India,
[email protected] [email protected] [email protected]

Abstract— A photoplethysmogram (PPG) is a plethysmogram the PPG signal with which blood pressure and heart rate
which is optically obtained, a volumetric measurement of an monitored continuously. Blood pressure is the measurement of
organ. Two PPG signals are used here to estimate the blood force applied to the blood vessels during blood circulating
pressure and heart rate. Blood pressure is the pressure exerted which is decreases as it moves away from the heart through
by circulating blood upon the walls of blood vessels. Nowadays
arteries and capillaries, and toward the heart through veins.
the standard blood pressure is measured by using cuff based
digital devices. In cuff based method, there will be an For each heartbeat, blood pressure measurement varies
uncomfortable to the patient and cannot measure the blood between systolic and diastolic pressures. The highest pressure
pressure in a continuous manner. Blood pressure is in terms of occurs when blood is travels through the arterial circulation by
the systolic (maximum) pressure over diastolic (minimum) the contraction of the heart which knows as the ‘systolic’
pressure and is measured in millimeters of mercury. Pulse blood pressure (SBP), while ‘diastolic’ blood pressure (DBP)
Transit Time (PTT) is playing the main role here. PTT is the measurement is taken during the heart relaxes when blood is
time between peaks of the two PPG signal. Heart rate can be travels through the arterial circulation by the contraction of the
measured by counting the number of peaks for sixty seconds. heart which knows as the ‘systolic’ blood pressure (SBP),
Blood pressure will be vary based on situation, activity, and
while diastolic while ‘diastolic’ blood pressure (DBP)
disease states. Heart rate is the speed of the heartbeat measured
by the number of contractions of the beat per minute (bpm). measurement is taken during the heart relaxes between beats
Threshold method is used to find the total number of peaks. The when the pressure in the arterial circulation falls to its lowest
signals are need to be filtered to remove any noises. A low pass level. Blood pressure is most commonly measured via a
filter is used here to remove the noises from the PPG signals. sphygmomanometer which consists of a combination of cuff,
Then the Pulse Transit Time (PTT) is determined by detecting inflating bulb with a release valve and a manometer. A
the peak location, peak time and peak time difference using manometer is a device which historically used the height of a
LabVIEW. Using linear regression analysis in MATLAB the column of mercury to reflect the circulating pressure.
blood pressure is obtained. GSM is used here to send the Sphygmomanometer has been the “gold standard” in non-
averaged PTT values.
invasive measurement for over 100 years.
Keywords— Physiological Parameters, Pulse Transit Time, Today blood pressure values are still reported in millimeters
Photoplethysmograpghy, Blood Pressure, Heart rate, GSM. of mercury (mmHg). Nanotechnology promises further a
better innovation in medical instrumentation which
optimistically sustains human healthiness.
I. INTRODUCTION
Johan et al proposed Body Sensor Network for
Technology have been improved for health monitoring Mobile Health Monitoring [1]. Here, Blood pressure
systems, which are capable of communicating wirelessly. measurement techniques are classified into two namely direct
Power consumption of the system is very low. Development and indirect method. B.R. Teja discussed about pulse transit
of such kind of devices are essential as they help in the time calculation from PPG. A hardware unit was instrumented
detection of abnormalities. Because of continuous monitoring, and software algorithm was implemented in that paper [2].
this systems are also used for emergency purpose in the Mahdi Boloursaz Mashhadi et al delivered tracking the heart
medical field and provides the user with an alert or warning rate using wrist type PPG signals [3]. Here, a novel algorithm
signal. Cost effective systems are being developed for is proposed consists of motion artifact cancellation and
continuous monitoring of the patient’s health condition. Blood spectral analysis. Xiaochuan et al suggested Detection of
pressure and pulse rate are the important parameters need to secondary peak of PPG signal for the measurement of blood
be continuously monitored. Blood pressure is measured pressure without cuff [4]. In this paper, secondary peak is used
mostly by using cuff based devices. The proposed system for the blood pressure measurement. Direct techniques or
consists of IR sensors which is used to continuously acquire invasive techniques gives continuous and reliable data about
Department of Electrical Technology, Karunya University IEEE ISBN No.978-1-5090-4527-3
311

Authorized licensed use limited to: University of Electronic Science and Tech of China. Downloaded on October 11,2024 at 01:48:32 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of IEEE International Conference on Innovations in Electrical, Electronics, Instrumentation and Media Technology
ICIEEIMT 17

the absolute vascular pressure from sensors placed inside sixty seconds will give the heart rate. The PTT values will be
blood stream. Biao Sun et al proposed Heart Rate Monitoring send through GSM for real time monitoring of the blood
using Asymmeric Least Squares Spectrum Subtraction using pressure.
PPG [5]. Here, a novel method called SPECTRAP is proposed
for accurate motion-tolerant estimation of Heart rate. Sangeeta PPG My Filtering
Bagha et al delivered Measurement of SpO2 and Pulse rate Sensors DAQ
using PPG signal [6]. Measurement of saturation of oxygen
and pulse rate is defined in this paper. The indirect methods or
non-invasive techniques consist of simple instruments and
cause little discomfort but intermittent and less information.
Devin B. McCombie et al proposed Adaptive Blood pressure
estimation from PPG sensors [7]. In this paper Pulse Wave PPG Peak Peak time Peak time
Velocity is estimated using circulatory waveform signals Detection Calculation difference
derived from multiple PPG sensors. Parry Fung et al described estimation
Non-invasive Continuous Blood pressure measurement by
pulse transit time [8]. C.C.Y. Poon et al proposed Cuff-less
Heart rate
and Non-invasive measurements of arterial blood pressure by Blood pressure
pulse transit time [9]. However, the accuracy gap between the estimation
estimation
invasive and the non-invasive methods, and has been
narrowing with the increasing computational power which can
process signal algorithms in terms of nanosecond. Monitoring System GSM Remote
station
Blood pressure is a function of cardiac output, which
is the amount of per cycle of blood volume output. An Fig. 1. System Overview for Signal Acquisition and
increased flow rate of blood will cause the blood pressure to Parameter Measurement
rise which was noted earlier. X.F. Teng et al evaluated a PTT B. PPG Sensor
based cuffless estimation of blood pressure in continuous
The sensor implemented is made up of two parts, a
manner[10]. Thus, there exists a linear relationship between
transmitter and a receiver. The transmitter consists of LEDs
the blood pressure value and the rate at which the blood
travels in the arteries. This flow rate can be calculated if the (IR and Red) which serve as the light sources at different
pulse wave velocity is known. wavelengths and the receiver a photo detector such as a
Photodiode. Two types of sensing exists namely reflective and
transmissive, where in reflective sensing, the transmitter and
Pulse transit time is the time interval for the arterial
receiver are located on the same place of the tissue and in
pulse pressure wave to travel from a valve called aortic where
transmissive sensing, the transmitter and receiver are situated
it is ejected to the peripheral site from the left ventricle. This
peripheral site can be anywhere along an artery. The pulse can on opposite sides of the tissue. Reflective sensing was
be felt the most at the artery called brachial. It is the major implemented due to the fact that transmissive sensors can only
blood vessel of the upper arm, for this reason it is the artery be fitted to thinner parts of the body. The other disadvantage
of using a transmissive based sensor is the fact that when there
that gets occluded when measuring blood pressure using the
is a drop in perfusion an accurate measurement cannot be
traditional mercury sphygmomanometer.
guaranteed.
II. MATERIALS AND METHODS
The pulse oximetry was used to measure a PPG
A. System Overview for PPG Signal Acquisition signal which will be used in order to derive both heart rate and
The required PPG signals are collected from the PPG this sensing nodes hardware is also used to acquire the signal
sensors. One signal from the fingertip and other signal from for calculating the blood pressure, which will be explained at a
the wrist is given to the DAQ. The DAQ is used for digital later stage. For the implementation of the receiver a
acquisition of the two analog input signals. Then the signals photodiode was used rather that a phototransistor since it
are given to LabVIEW. In LabVIEW a low pass filter is used yields a much faster response. The measured (reflected) light
to remove the noises in the PPG signal. The frequency range induces a current in the photodiode which is converted to a
of PPG signal is 0.5 - 5 Hz. Threshold method is used for peak voltage by the transimpedance amplifier. The photodiode will
detection. First the time of the two peaks is calculated and the detect the reflected light however the light absorbed by the
distance between the two peaks is estimated. The time haemoglobin needs to be used to determine the oxygen levels.
between the two peaks of the PPG signal is called Pulse Therefore, the signal needs to be inverted to produce the
Transit Time. The blood pressure is estimated through absorbed light this is implemented in the transimpedance
regression analysis. The number of peaks in the signal for amplifier stage.

Department of Electrical Technology, Karunya University IEEE ISBN No.978-1-5090-4527-3

312

Authorized licensed use limited to: University of Electronic Science and Tech of China. Downloaded on October 11,2024 at 01:48:32 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of IEEE International Conference on Innovations in Electrical, Electronics, Instrumentation and Media Technology
ICIEEIMT 17

C. Pulse Transit Time to remove the noises from the PPG signal at a cut-off
frequency of 5Hz. Sampling frequency is 1KHz.
Pulse transit time (PTT) is the time difference between the two
PPG signals. One signal is taking from the wrist and the other
is taking from the fingertip. It is the time taken by a pulse
wave to propagate from heart to a particular point on the body
where the reading is taken. The two PPG sensors are set up in
such a way that one is leading and the other is lagging. Both
sensors measure PPG signals, however due to their location
and a relevant distance from the heart the one signal is leads
the other in a time domain measurement. Pulse transit time
can be find by calculating the peak time period of the two
signals.

D. Estimation of Blood Pressure using PTT

The PTT decreases as the Blood Pressure increases,
while the Blood Pressure increases as PTT decreases. A linear
regression equation using PTT as an input variable can be
derived establishing the compensation coefficient according to Fig. 2. Experimental Setup
correlation, after the Blood Pressure and PTT are measured
using the inverse relation between the Blood Pressure and The peak detector is used here to detect the peak locations of
PTT. The systolic and the diastolic Blood Pressure can be the two PPG signals. A threshold value is given to detect the
estimated using PTT only after the linear regression equation peaks of the two signals. For finger PPG the threshold value is
estimating the Blood Pressure is derived. The Blood Pressure 0.5 and for wrist PPG the threshold value is 1.5. The locations
measurement method using PTT has the advantage of of the two PPG signals is multiplied with dt respectively. dt
providing the convenient real-time monitoring of the Blood returns the time interval in seconds between data points in the
Pressure without the cuff. waveform. The multiplied output is then added with t0. T0
returns the trigger time of the waveform. The two added ouput
E. Communication Methods is subtracted, the subtracted output is the time difference
In this paper, GSM is used to send the PTT values to between the two PPG signals. i.e Pulse Transit Time (PTT).
the remote station. GSM is a mobile communication modem it
is stands for global system for mobile communication (GSM).
GSM is an open and digital cellular technology used for
transmitting mobile voice and data services operates at the
850MHz, 900MHz, 1800MHz and 1900MHz frequency
bands. GSM system was developed as a digital system using
time division multiple access (TDMA) technique for
communication purpose. A GSM digitizes and reduces the
data, then sends it down through a channel with two different
streams of client data, each in its own particular time slot. The
digital system has an ability to carry 64 kbps to 120 Mbps of Fig. 3. Input PPG Signal
data rates. There are various cell sizes in a GSM system such
as macro, micro, pico and umbrella cells. Each cell varies as The PTT values are shown in the above figure. Two waveform
per the implementation domain. The coverage area of each charts shows the PPG signals taken from the finger and the
cell varies according to the implementation environment. wrist respectively. Locations and amplitudes of the signals are
also obtained. The PTT values are shown in milliseconds. The
III. RESULTS AND DISCUSSION PTT values must be averaged. Time of the two PPG signals is
The experimental setup consists of two IR sensors of also obtained. A milliseconds of change in peak time between
reflective type one at the finger and other at the wrist. Two the two signals is observed.
signals are taken at a same time from the two places. A DAQ
is connected to acquire the analog input voltages and to Due to the location and relevant distances from the heart the
convert into digital. A low pass filter is used in the LabVIEW one signal leads the other in a time domain measurement. This
time is measured using the program in the LabVIEW and used

Department of Electrical Technology, Karunya University IEEE ISBN No.978-1-5090-4527-3

313

Authorized licensed use limited to: University of Electronic Science and Tech of China. Downloaded on October 11,2024 at 01:48:32 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of IEEE International Conference on Innovations in Electrical, Electronics, Instrumentation and Media Technology
ICIEEIMT 17

in the calculation and prediction of blood pressure. A very The table 1 shows the comparison of standard blood pressure
small amount of time difference only will occur between the and estimated blood pressure. The error difference between
signals. The signal which is taking from the wrist is always the standard blood pressure and estimated blood pressure is ±5
leading because it is nearer than the fingertip to the heart. The or ±7. The accuracy can be increased based on the sensitivity
signal which is taking from the fingertip is always lagging. of the sensor.

Table. 2. Standard Heart rate and estimated Heart Rate

Standard Estimated
S. No Patient name Pulse rate Pulse rate
(BPM) (BPM)

1 Patient 1 65 72

2 Patient 2 96 96

3 Patient 3 75 84

4 Patient 4 85 72

5 Patient 5 77 60

Fig. 4. PTT Calculation from PPG signal

The Blood pressure is finally measured by the Pulse transit The table 2 shows the comparison of Standard Pulse rate and
time by predicting it in the linear regression analysis. The Estimated Pulse rate. The error difference between the
following table shows the measurement of blood pressure standard pulse rate and estimated pulse rate is ±7 or ±9.V
through the regression principles and compared with standard
IV CONCLUSION
blood pressure measurement device.
The overall results achieved are very closely related
Table. 1. Standard Blood pressure and Estimated Blood to the standard results, this means that the final
pressure implementation of the project works as intended. More
number of chances are there of getting the wrong results if the
patients change their hand position. In case if the patients want
Standard blood Estimated blood to change their hand position means at that time also the
pressure pressure sensors should be able to sense the signal. Because of this is a
S. Patient (mmHg) (mmHg) cuffless and continuous measurement technique it will be an
NO name easy and advanced way of measuring the blood pressure and
pulse rate.
Systolic Diastolic Systolic Diastolic
Pressure Pressure Pressure Pressure Acquisition of two PPG signals at the same time was
done. Pulse Transit Time was measured successfully for
Patient estimating the blood pressure. Blood pressure was estimated
1 101 65 114 73
1 successfully from the measured PTT through linear regression
Patient analysis. Pulse rate was estimated by calculating the total
2 118 81 114 73
2 number of peaks for sixty seconds. A GSM module was also
Patient interfaced to send the averaged PTT value. This will be a
3 119 75 110 71
3 promising way to find continuous blood pressure monitoring
Patient in home as well as in hospitals without using cuff. Moreover it
4 107 68 112 72
4 is a non-invasive technique which helps in determining Blood
Patient Pressure more accurately in all ranges. Future work will
5 120 78 117 74
5 include more data analysis from patients to make the system
more efficient. Sensors will be designed of more sensitivity on
acquiring the signal at the time of disturbances also.

Department of Electrical Technology, Karunya University IEEE ISBN No.978-1-5090-4527-3

314

Authorized licensed use limited to: University of Electronic Science and Tech of China. Downloaded on October 11,2024 at 01:48:32 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of IEEE International Conference on Innovations in Electrical, Electronics, Instrumentation and Media Technology
ICIEEIMT 17

REFERENCES

[1] Johan Wannenburg and Reza Malekain, (2015) “Body Sensor Network [6] S. Bagha and L. Shaw, (2011) “A real time analysis of PPG signal for
for Mobile Health Monitoring, a diagnosis and Anticipating System,” measurement of SpO2 and pulse rate,” Int. J. Comput. Appl., vol. 36, no.
IEEE Sensors Journal, Vol. 15, No. 12, pp 6839-6852. 11, pp. 45–50.
[2] B. R. Teja, (2012) “Calculation of blood pulse transit time from PPG,” [7] Devin B. McCombie, Andrew T. Reisner, H. Harry Asada, (2006), “
Ph.D. dissertation, Dept. Biotechnol. Med. Eng., National Institute of Adaptive Blood pressure estimation from wearable PPG sensors using
Technology, Rourkela, India. peripheral artery pulse wave velocity measurements and multi-channel
[3] Mahdi Boloursaz Mashhadi, Ehsan Asadi, Mohsen Eskandari, Shahrzad blind identification of local arterial dynamics.” International Conference
Kiani, (2016) “Heart Rate Tracking Using Wrist-Type of the IEEE Engineering in Medicine and Biology, pp 3521-3524.
Photoplethysmographic(PPG) Signals during Physical Exercise with [8] P. Fung, G. Dumont, C. Ries, C. Mott, and M. Ansermino,(2004)
Simultaneous Accelerometry,” IEEE Signal Processing Letters, Vol. 23, “Continuous noninvasive blood pressure measurement by pulse transit
No. 2, pp 227-231. time,” in Proc. Annu. Int. Conf. IEEE Eng. Med. Biol. Soc., vol. 1. San
[4] Xiaochuan He, Rafik A. Goubran, Xiaoping P.Liu, (2014) “ Secondary Francisco, CA,USA, pp. 738–741.
Peak Detection of PPG signal for Continuous Cuffless Arterial Blood [9] C.C.Y. Poon and Y. T. Zhang, (2005) “Cuff-less and Non-invasive
Pressure Measurement,” IEEE Transactions on Instrumentation and Measurements of Arterial Blood Pressure by Pulse Transit Time,” IEEE
Measurement, Vol. 63, No. 6, pp 1431-1439. Engineering in Medicine and Biology, pp 5877-5880.
[5] Biao Sun and Zhilin Zhang, (2015) “Photoplethysmography-Based [10] X. F. Teng and Y.T. Zhang, (2006) “An Evaluation of a PTT based
Heart Rate Monitoring Using Asymmetric Least Squares Spectrum method for Non-invasive and cuffless estimation odf arterial blood
Subtraction and Bayesian Decision Theory.” IEEE Sensors Journal, Vol. pressure,” International conference of the IEEE Engineering in Medicine
15, No. 12, pp 7161-7168. and Biology.

Department of Electrical Technology, Karunya University IEEE ISBN No.978-1-5090-4527-3

315

Authorized licensed use limited to: University of Electronic Science and Tech of China. Downloaded on October 11,2024 at 01:48:32 UTC from IEEE Xplore. Restrictions apply.