Islamic University Sample Question Solution 2019 (Biomedical Signal Processing)

Sample Question
• Set 1: Biomedical signal, source of signal, classification of signal.
• set 2:[Heart Physiology]Describe Hearts electrical conduction
system.Action potentials
• Set 3:[ECG components]:P wave, PQ segment, QRS, ST , T wave.
Definitions:EEG,EMG,EOG signals
• Set 4:[HRV]:Time/Frequency domain,Stastical,Geometric,Spectral
components(long,shortseries),StasticalTDHRV GeometricTDHRV
• SET 5: z transformation, LTI system, ROC, Zeros, Poles, math page 138
(2.16)
• Set 6: Entropy (approximate,sample,corrected,parametric)
• Set 7: Model identification,AR model, Auto variance/correlation cross
correlation
• Set 8: [part one] acceleration signals, Activity count, methods of activity
count, application of accelerators.
• Set 8: [part 2] Sleep stage classification, physical Activity recognition.

1Q. Definition Biomedical signal?
Biomedical signal definition: The signal that conveys biological information about
the state or behavior of some living objects. Biomedical signals are used to realize
the underlying physiological mechanisms of certain biological system or event.
Example: Human blood pressure, ECG, EEG, etc.
Types: Biomedical signals can be categorized according to the source of origination,
number of channels, dimensionality, model, and nature as illustrated

Q. Biomedical Signal Classification?
Q. Source of signal?
Signal source: Biomedical signals originate from different sources: auditory system, nervous system,
cardiovascular system, endocrine system, circulatory system, musculoskeletal system, vision system,
gastrointestinal system, and respiratory system.
Number of Channels: Based on the required number of channels to acquire a specific biomedical signal, it can
be one channel to display the pulse wave, three channels to display the accelerometer data, or multichannel
with the electroencephalography (EEG) signals.
Model: Based on the signal models (analysis approach), the biomedical signals can be deterministic or
stochastic. The deterministic biomedical signal is either periodic (sinusoidal or complex) or non-periodic
(transient), which is predictable. The stochastic biomedical signals can be stationary or nonstationary.
Deterministic vs Stochastic Signals: Deterministic signals can be described by some mathematical expressions
(or formulas). Stochastic signal is a signal with some uncertainty

Q2. (Heart Physiology) Describe Hearts electrical conduction system?
Action potentials?
• The heart has four chambers: two atria and two ventricles.
• The right atrium receives oxygen-poor blood from the body and pumps it to the right ventricle.
• The right ventricle pumps the oxygen-poor blood to the lungs.
• The left atrium receives oxygen-rich blood from the lungs and pumps it to the left ventricle.
• The left ventricle pumps the oxygen-rich blood to the body.
The coronary arteries are the blood vessels (arteries) of coronary circulation, which transports oxygenated blood
to the actual heart muscle.

3Q. Definitions of ECG,EEG,EMG,EOG signals.
EEG signals: EEG measures voltage fluctuations resulting from ionic current within
the neurons of the brain. Clinically, EEG refers to the recording of the brain's
spontaneous electrical activity over a period of time, as recorded from multiple
electrodes placed on the scalp.
EMG signals: The EMG signal is a biomedical signal that measures electrical
currents generated in muscles during its contraction representing neuromuscular
activities. The nervous system always controls the muscle activity
(contraction/relaxation).
EOG signals: Electrooculography (EOG) is a technique for measuring the corneo-
retinal standing potential that exists between the front and the back of the human
eye. The resulting signal is called the electrooculogram. Primary applications are in
ophthalmological diagnosis and in recording eye movements.
ECG signals: Electrocardiography is the process of producing an electrocardiogram
(ECG or EKG), a recording – a graph of voltage versus time – of the electrical activity
of the heart using electrodes placed on the skin. ... This orderly pattern of
depolarization gives rise to the characteristic ECG tracing.

Q. ECG components of P wave, PQ segment, QRS, ST , T wave.
There are three main components to an ECG: the P wave, which
represents the depolarization of the atria; the QRS complex, which
represents the depolarization of the ventricles; and the T wave, which
represents the repolarization of the ventricles.

Q. Definition of HRV? HRV Task Force Goals?
Heart Rate Variability: The variations in the intervals between successive heart beats or variation in
the instantaneous heart rates
• There is a significant relationship betn autonomic nervous system and cardiovascular mortality
• HRV is one of the most prominent markers of autonomic activity
HRV Task Force Goals: The significance and meaning of many different measures of HRV are more
complex than generally appreciated. So, In 1996
The European Society of Cardiology and the North American Society of Pacing and
Electrophysiology Constitute a Task Force
Goals:
1. Standardize nomenclature and develop definitions of terms
2. Specify standard methods of measurement
3. Define physiological and pathophysiological correlates
4. Describe currently appropriate clinical applications
5. Identify areas for future research
Different method HRV
Measurement:
• Time domain
• Frequency domain
• Statistical
• Geometric

4Q. Time domain HRV
Time Domain HRV Measurements
• meanNN-The mean of N-N heart beat intervals
• meanHR- the mean of instantaneous heart rates
• SDNN-Standard deviation of all N-N intervals
• SDANN-SDNN of the avg of NN intervals in all 5-mins segments
• RMSSD-square root of the sum of the squares of differences betn
adjacent NN intervals
• NN50 count- No. of pairs of adjacent NN intervals differing by more
than 50 ms
• PNN50-NN50 count/no. of all NN intervals

Q. Frequency domain
Frequency Domain HRV Measurements
Power Spectral Density (PSD) analysis provides
the basic information about how the power
(or variance) of a series is distributed as a
function of time (t)
PSD methods:
1. Nonparametric (FFT),
2. Parametric (AR, MA, ARMA)
Advantages of Nonparametric
• Very simple algorithm
• High processing speed
Advantages of Parametric:
• Smoother spectral components
that can be distinguished
• An accurate estimation of PSD
even on a small series
Limitations: Stationarity

Q. Statistical TD HRV Measurements
Statistical measures are used for longer period of time series (24 hrs or more)
1. From direct NN intervals
2. Derived from the difference of NN intervals
 These variables may be derived from the analysis of total ECG recording
 From smaller segments of the recording which allows comparisons of the HRV during varying
activities: rest, sleep, active,
• SDNN- reflects all the cyclic components responsible for variability in the period of recording
It may calculate from 24 hrs record (or shorter period)
As the monitoring period decreases, SDNN estimates shorter
Other Statistical measures
• SDANN-estimates changes in HR due to cycles longer than 5 mins
Most Commonly Used
• RMSSD, NN50 count, PNN50
All these measurements of short term variations estimates HF variations in HR, and thus are highly correlated

Q. Geometric TD HRV Measurements
The RR( i.e., NN) series is converted into a geometric shape such as sample density distribution of RR series, and a simple formula is used that judges the
variability on the basis of the geometric pattern
Three General Approaches:
1. A basic measurement of the geometric pattern (width of the distribution histogram)
2. The pattern is approximated by mathematically defined shape (Triangle)
3. The shape is classified into different pattern based categories (Linear, triangular, elliptic of
Lorenz plots)
N.B. In common, bin length 8 ms (i.e., 1/128 sec)
Lorenz plot
A scatter plot that shows the RR interval as a function of preceding RR interval (i.e., Rri vs Rri+1)
• This plot might be used for visualizing the variability of heart rate
• The assessment of HRV using this plot might be superior to conventional HRV measures (mean,
sdnn, sdann, etc)
• A precise numeric evaluation of the images of Lorenz plots is possible (Katerina Hnatkova)
• Plots that are very compact result in a sharply failing density function, while plots that are more
diffuse lead to a flat density function

Q. Spectral Components for Short-term & Long Series
Short-term
• Three main spectral components: VLF, LF, HF are distinguished in a spectrum calculated from
short recordings of 2-5 mins
• The VLF assessment from short series can be affected due to trend or baseline, and hence should
be avoided.
• The VLF, LF, and HF is usually made in absolute values of the power (ms2)
• LF and HF may also be measured in normalized units i.e., the relative value of each power
component in proportion to the total power minus the VLF component
• The normalized LF and HF emphasize the controlled and balanced behavior of two branches of
ANS
Long Series
• The spectral analysis of long-term recordings includes ULF (0-0.003 Hz) component in addition to
VLF, LF, and HF components
• The question of ‘‘stationarity” frequently arises with long series
• If mechanisms responsible for heart period modulations of a certain frequency remain unchanged
during the whole recordings, the corresponding frequency component of HRV can be used as a
measure of these modulations.

5Q. z-transformation ? LTI system ?
In mathematics and signal processing, the Z-transform converts a discrete-time
signal, which is a sequence of real or complex numbers, into a complex frequency-
domain representation. It can be considered as a discrete-time equivalent of the
Laplace transform.
• Concept of Z-Transform and Inverse Z-Transform
X(Z)|z=ejω=F. T[x(n)].
LTI system: Linear time-invariant systems (LTI systems) are a class of systems used
in signals and systems that are both linear and time-invariant. Time-invariant
systems are systems where the output does not depend on when an input was
applied. These properties make LTI systems easy to represent and understand
graphically.

Q. ROC(Region of Convergence)
• ROC contains strip lines parallel to jω axis in s-plane.
•If x(t) is absolutely integral and it is of finite duration, then ROC
is entire s-plane.
•If x(t) is a right sided sequence then ROC : Re{s} > σo.
•If x(t) is a left sided sequence then ROC : Re{s} < σo.
•If x(t) is a two sided sequence then ROC is the combination of
two regions.

6Q. Approximate Entropy? Sample Entropy? Corrected Conditional
Entropy
Approximate Entropy: In statistics, an approximate entropy (ApEn) is a
technique used to quantify the amount of regularity and the unpredictability
of fluctuations over time-series data. Regularity was originally measured by
exact regularity statistics, which has mainly centered on various entropy
measures.
Sample entropy (SampEn) is a modification of approximate entropy (ApEn),
used for assessing the complexity of physiological time-series signals,
diagnosing diseased states. SampEn has two advantages over ApEn: data
length independence and a relatively trouble-free implementation.
Corrected Conditional Entropy
• Depends on three parameters: (N), L (=M-1) and »
• Measures: entropy rate of a symbolic sequence derived by discretizing the
signal

7Q. Model identification?
• In practice, we do not know the coefficients𝑎𝑖, the model order𝑃and the
parameters of 𝑒𝑘(𝜇𝑒, 𝜎𝑒2)
• We have only a set of data points𝑌={𝑦1,𝑦2,…,𝑦𝑁}
• Model identification aims to determine the model parameters from the
data model fitting
• Many different approaches have been invented
•The most popular one minimizes the variance of the prediction error 𝑣𝑎𝑟[𝜖
]computed on the data
𝐽𝑎,𝜇𝑒,𝜎𝑒2,𝑃𝑌=1𝑁 𝑘=1𝑁𝑌𝑘− 𝑦𝑘|𝑘−12
•𝜇𝑒 can be set to 0 subtracting the sample mean of 𝑌from each𝑌𝑖(ergodic
process)
•𝑎can be estimated by solving a linear system of equations(convex
optimization problem with an unique solution)
•𝜎𝜖2is𝐽when using the best coefficients found(when data is generated by the
model 𝐽=𝜎𝑒2)

Q. AR model ? Auto covariance ? Auto correlation? Cross
correlation?
AR model: An AR(p) model is an autoregressive model where specific lagged
values of yt are used as predictor variables. Lags are where results from one
time period affect following periods. The value for “p” is called the order.
Auto covariance: In probability theory and statistics, given a stochastic
process, the auto covariance is a function that gives the covariance of the
process with itself at pairs of time points. Auto covariance is closely related
to the autocorrelation of the process in question.
Autocorrelation: Autocorrelation, also known as serial correlation, is the
correlation of a signal with a delayed copy of itself as a function of delay.
Informally, it is the similarity between observations as a function of the time
lag between them.
Cross correlation is a measurement that tracks the movements of two
variables or sets of data relative to each other. In its simplest version, it can
be described in terms of an independent variable, X, and two dependent
variables, Y and Z.

8Q. Acceleration signals?
Acceleration signals: An accelerometer is a device that measures the vibration, or
acceleration of motion of a structure. The force caused by vibration or a change in motion
(acceleration) causes the mass to "squeeze" the piezoelectric material which produces an
electrical charge that is proportional to the force exerted upon it. Since the charge is
proportional to the force, and the mass is a constant, then the charge is also proportional
to the acceleration.
• Amplification to increase measurement resolution and improve signal-to-
noise ratio
• Current excitation to power the charge amplifier in IEPE sensors
• AC coupling to remove DC offset, increase resolution, and take advantage
of the full range of the input device
• Filtering to remove external, high-frequency noise
• Proper grounding to eliminate noise from current flow between different
ground potentials
• Dynamic range to measure the full amplitude range of the accelerometer

Q. Applications of Accelerometer signals
Accelerometers are used to measure the motion and vibration of a structure that is
exposed to dynamic loads. Dynamic loads originate from a variety of sources
including:
•Human activities – walking, running, dancing or skipping
•Working machines – inside a building or in the surrounding area
•Construction work – driving piles, demolition, drilling and excavating
•Moving loads on bridges
•Vehicle collisions
•Impact loads – falling debris
•Concussion loads – internal and external explosions
•Collapse of structural elements
•Wind loads and wind gusts
•Air blast pressure
•Loss of support because of ground failure
•Earthquakes and aftershocks

Q. Methods of activity count
• most HAR solutions have been developed using artificial intelligence
methods throughvarious machine learning techniques, including
shallow (e.g., Sensors 2019, 19, 3213; doi:10.3390/s19143213
www.mdpi.com/journal/sensors Sensors 2019, 19, 3213 2 of 28
Support Vector Machine (SVM), Decision Tree, Naive Bayes, and KNN)
and deep algorithms (e.g., Convolutional Neural Network (CNN),
Recurrent Neural Network (RNN), Restricted Boltzmann Machine
(RBM), Stacked Autoencoder (SAE), Deeply-Connected Network
(DFN), and Deep Belief Network (DBN))

Q. Sleep stage classification ?
• An alternative technique for sleep stages classification based on
heart rate variability (HRV) was presented in this paper. The simple
subject specific scheme and a more practical subject independent
scheme were designed to classify wake, rapid eye movement (REM)
sleep and non-REM (NREM) sleep.
Highlights
• We classify different sleep stages with 41 HRV features
• We evaluate the importance of every feature for sleep staging.
• The classification performance is in prior to previous relevant studies.
• Some new proposed features perform even better than the conventional ones.
• The first 10 features could lead to considerable results compared to all features.

Q. Physical activity recognition
Physical activity (PA) is any bodily movement worked by skeletal muscles that requires more energy
expenditure than resting, such as walking, running, swimming, or aerobic exercise and strength training.
Getting proper physical activity throughout the day can lower the risk of type 2 diabetes, cardiovascular
disease, stroke, and obesity. However, most people do not do enough physical activity, because it is not easy to
be measured. Therefore, accurate tracking and monitoring of PA under free-living conditions is of significant
importance to cultivate scientific living habits and improve an individual’s health.
Multi-Sensor Measurement Platform
Multi-Sensor Monitoring System
Based on the WIMS:
WIMS System Design and Realization:
• 1. Hip Unit: one tri-axial accelerometer ADXL345 worn at the hip, to measure the body motions that
characterize the degree of PA of the lower part of the body;
• 2. Wrist Unit: one tri-axial accelerometer ADXL345 worn on the wrist, to measure the arm and hand motions
that characterize the PA of the upper part of the body;
• 3. Abdominal Unit: one ventilation sensor made of piezoelectric crystal wrapped around the abdomen, for
measuring the expansion and contraction resulting from the subject’s respiration

Islamic University Sample Question Solution 2019 (Biomedical Signal Processing)

More Related Content

What's hot (20)

Similar to Islamic University Sample Question Solution 2019 (Biomedical Signal Processing) (20)

More from Rakibul Hasan Pranto (7)

Recently uploaded (20)

Islamic University Sample Question Solution 2019 (Biomedical Signal Processing)