Mas S Mohktar

Email: [email protected]
Phone (office): 0379677681
 
At the end of the course, students are able to:
1. Evaluate the concepts and theories of statistics used in
biomedical engineering (PO2)
2. Apply statistical package to analyze biomedical
engineering problems (PO5)
3. Formulate statistical analysis related to the management
in biomedical engineering (PO12)
 PO 2: Identify, formulate, research, analyze and reach
substantiated conclusions along with recommendations for
complex biomedical engineering problems, using principles
of mathematics, natural sciences and engineering sciences.
 PO 5: Utilize systematic approach, resources, modern
engineering and IT tools, with full understanding of their
limitations, to complex biomedical engineering activities.
 PO 12: Apply knowledge of engineering management
and financial principles.
 Continuous assessments : 50%
o Entry Quiz (5%) – CO1
o Mid – Sem Test (15%) – CO1
o Exercise (5%) – CO1
o 1 Group Report (15%) – CO2, CO3
o 1 Presentation (10%) – CO2, CO3
 Final Exam: 50% - CO1, CO2
 Probability and probability distribution
 Sampling Distributions and Point Estimation of Parameters
 Statistical Intervals
 Tests of Hypotheses
 Statistical Inference for Single sample
 Statistical Inference for Two Samples
 Simple Linear Regression and Correlation
 Multiple Linear Regression
 Field of statistics:
o The study and use of theory and methods for the analysis of data
o Deals with the collection, presentation, analysis and use of data to
make decisions, solve problems, and design products and
processes

 Statistics is the science of data.

 Biostatistics
o Branch of applied statistics directed toward applications in the
health sciences and biology
o Statistical methods are more heavily used in health applications
(e.g., survival analysis, longitudinal data analysis)
 Why we need biostatistics?
1. To choose the best therapy, a physician must compare the
prognosis, or future course, of a patient under several therapies. A
therapy may be a success, a failure, or somewhere in between; the
evaluation of the chance of each occurrence necessarily enters
into the decision
2. In designing and planning medical care facilities, planners take
into account differing needs for medical care. All of the uncertainty
associated with the future health of a population and its future
geographic and demographic patterns should be taken into
account.
 Are observations of random variables made on the
elements of a population or sample
 Are the quantities (numbers) or qualities (attributes)
measured or observed that are to be collected and/or
analysed
 Qualitative variables have values that are intrinsically
nonnumeric (categorical)
o E.g., Cause of death, race, gender, severity of pain (mild, moderate,
severe)
o Generally are either nominal or ordinal
o Can be reassigned numeric values (e.g., male=0, female=1)
 Quantitative variables have values that are intrinsically
numeric
o E.g., survival time, systolic blood pressure, height
o Can be further subdivided into discrete and continuous variables
o Discrete variables have a set of possible values that is either finite
or count ably infinite
o Continuous variable has a set of possible values including all values
in an interval of the real line
 Collecting engineering data
o Retrospective study based on historical data
o Observational study
o Designed experiment
 Uses either all or a sample of the historical data from some
period of time

 Example:

“Misdiagnosis of the vegetative state: retrospective study in a

rehabilitation unit”
BMJ 1996;313:13
Keith Andrews, Lesley Murphy, Ros Munday, and Clare Littlewood

Method
The medical, occupational therapy, and clinical psychology records of all
patients admitted to the rehabilitation unit for profound brain damage
between 1992 and 1995 were examined retrospectively. Patients with a
referral diagnosis of the vegetative state due to acute onset brain damage
were included.
 Observe the process or population during a period of routine operations
 No intervention
 An observational study collects data from an existing situation
 The data collection does not intentionally interfere with the running of
the system
 Example:
Mortality after all major types of osteoporotic fracture in men and women: an
observational study
The Lancet, Volume 353, Issue 9156, 13 March 1999, Pages 878–882
Jacqueline R Center, Tuan V Nguyen, Diane Schneider, Philip N Sambrook, John A
Eisman
A 5-year prospective cohort study in the semi-urban city of Dubbo, Australia, of all
residents aged 60 years and older (2413 women and 1898 men). Low-trauma
osteoporotic fractures that occurred between 1989 and 1994, confirmed by
radiography and personal interview, were classified as proximal femur, vertebral,
and groupings of other major and minor fractures. We calculated standardised
mortality rates from death certificates for people with fractures compared with the
Dubbo population.
 Where a cohort of people is followed for the occurrence or
nonoccurrence of specified endpoints or events or
measurements.
 Example:
Changes in Physical Fitness and All-Cause MortalityA Prospective Study of
Healthy and Unhealthy Men
JAMA. 1995;273(14):1093-1098
Steven N. Blair; Harold W. Kohl; Carolyn E. Barlow; Ralph S. Paffenbarger; Larry
W. Gibbons; Caroline A. Macera,
Design. —Prospective study, with two clinical examinations (mean interval
between examinations, 4.9 years) to assess change or lack of change in
physical fitness as associated with risk of mortality during follow-up after the
subsequent examination (mean follow-up from subsequent examination, 5.1
years).
 An experiment is a study in which an investigator
deliberately sets one or more factors to a specific level
 Laboratory experiment
o takes place in an environment (called a laboratory) where
experimental manipulation is facilitated
 Comparative experiment
o compares two or more techniques, treatments, or levels of a
variable
 Crossover experiment
o same experimental unit receives more than one treatment or is
investigated under more than one condition of the experiment
o the different treatments are given during non overlapping time
periods
 A clinical study is one that takes place in the setting of
clinical medicine
 A study that takes place in an organizational unit
dispensing health care—such as a hospital, psychiatric
clinic, well-child clinic, or group practice clinic—is a clinical
study
 Case studies or case series
o A descriptive account of interesting characteristics e.g. symptoms
observed in a single case subject with disease or in a sample of
cases
o Typically are unplanned and don’t involve any research hypotheses
o No comparison group
o Poor design, but can generate research hypotheses for subsequent
investigation
 Case-control study
o Conducted retrospectively by looking into past
o Two types of subjects included:
• cases = subjects with the disease/outcome of interest
• controls = subjects without the disease/outcome
o History of two groups is examined to determine which subjects were
exposed to, or otherwise possessed, a prior characteristic
o Association between exposure and disease then quantified
o Controls are often matched to cases based on similar
characteristics
 Cross-sectional studies
o Collect data from a group of subjects at one point in time
o Sometimes called prevalence studies due to their focus on a single
point in time
 Longitudinal studies
o collects information on study units over a specified time period
 Cohort studies
o Usually conducted prospectively forward in time
o A cohort is a group of people who have something in common at a
particular point in time and who remain part of the group through
time
o A cohort of disease free subjects are selected and their exposure
status evaluated at the start of the study
o They are then followed through time in order to observe who
develops disease
o Association between exposures risk factors and disease are then
quantified
1. A question or problem area of interest is considered
2. A study is to be designed to answer the question. The design of the
study must consider at least the following elements:
a) Identify the data to be collected. This includes the variables to be measured as
well as the number of experimental units, that is, the size of the study or
experiment.
b) An appropriate analytical model needs to be developed for describing and
processing data.
c) What inferences does one hope to make from the study? What conclusions
might one draw from the study? To what population(s) is the conclusion
applicable?
3. The study is carried out and the data are collected.
4. The data are analyzed and conclusions and inferences are drawn.
5. The results are used. This may involve changing operating procedures,
publishing results,or planning a subsequent study.
 Ethics
 Design of forms
o What Data Are to Be Collected?
o Clarity of Questions
o Pretesting of Forms and Pilot Studies
o Layout and Appearance
1. Validity checks
o Check that only allowable values or codes are given for answers to
the questions. For example; a negative weight is not allowed
o A simple extension of this idea is to require that most of the data
fall within a given range; range checks are set so that a small
fraction of the valid data will be outside the range and will be
“flagged”; for example, the height of a professional basketball
team center (who happens to be a subject in the study) may fall
outside the allowed range even though the height is correct
o By checking out-of-range values, many incorrectly recorded values
can be detected
2. Consistency checks
a) If more than one form is involved, the dates on these forms should
be consistent with each other (e.g., a date of surgery should
precede the date of discharge for that surgery).
b) Consistency checks can be built into the study by collecting crucial
data in two different ways (e.g., ask for both date of birth and age).
c) If the data are collected sequentially, it is useful to examine
unexpected changes between forms (e.g., changes in height, or
drastic changes such as changes of weight by 70%). Occasionally,
such changes are correct, but they should be investigated.
d) In some cases there are certain combinations of replies that are
mutually inconsistent; checks for these should be incorporated into
the editing and verification procedures
3. Missing forms
o In some case–control studies, a particular control may refuse to
participate in a study. Some preliminary data on this control may
already have been collected. Some mechanism should be set up
so that it is clear that no further information will be obtained for
that control. (It will be useful to keep the preliminary information so
that possible selection bias can be detected.) If forms are entered
sequentially, it will be useful to decide when missing forms will be
labeled “overdue” or “missing.”
 Bias
 Similarity in a Comparative Study
 Inference to a Larger Population
 Precision and Validity of Measurements
 Quantification and Reduction of Uncertainty
 BME jobs expected to grow 72%
 https://bme.ucdavis.edu/blog/bme-jobs-expected-to-grow-
72/