SECTION A
(2 × 10 = 20 | Very Short Answers)
a. Role of computers in biomedical signal processing
Computers help in acquisition, storage, filtering, analysis, visualization, and interpretation of biomedical signals with high speed and accuracy.
b. Electro-Retinography (ERG)
ERG is a diagnostic test that records the electrical response of the retina to light stimulation to assess retinal function.
c. Purpose of portable arrhythmia monitors
They continuously monitor heart rhythm over long durations to detect irregular heartbeats (arrhythmias) during daily activities.
d. What does ECG measure?
ECG measures the electrical activity of the heart during depolarization and repolarization of cardiac muscles.
e. Purpose of Huffman coding in data reduction
Huffman coding reduces data size by assigning shorter codes to frequently occurring symbols, improving storage efficiency.
f. Use of run-length coding
Run-length coding compresses data by representing consecutive repeated values as a single value and count.
g. Epilepsy transition
Epilepsy transition refers to the change in brain state from normal EEG activity to seizure activity.
h. Periodogram in EEG spectral analysis
A periodogram estimates the power spectral density of EEG signals to analyze frequency components.
i. Signal averaging in EP estimation
Signal averaging reduces random noise by averaging repeated responses, improving signal-to-noise ratio.
j. Adaptive noise cancelling
Adaptive noise cancelling removes unwanted noise by continuously adjusting filter parameters based on input signals.
SECTION B
(Attempt any THREE | 3 × 10 = 30 Marks)
2(a) Acquisition of biomedical signals & difficulties
Biomedical signal acquisition involves sensors/electrodes, amplification, filtering, analog-to-digital conversion, and recording.
Difficulties include low signal amplitude, noise interference, motion artifacts, electrode-skin impedance, baseline drift, and power line interference.
2(b) QRS detection methods in ECG signals
QRS detection identifies ventricular depolarization in ECG. Common methods include:
Threshold-based detection Derivative-based methods
Pan-Tompkins algorithm Wavelet-based detection
These methods improve heart rate estimation and arrhythmia detection.
2(c) Turning Point algorithm for data reduction
The Turning Point algorithm reduces data by selecting points where signal direction changes.
Key features: Preserves signal shape
Reduces storage requirements Simple and efficient
It removes redundant points while retaining diagnostic information.
2(d) Importance of neurological signal processing in EEG
EEG processing helps analyze brain activity for epilepsy, sleep disorders, and neurological diseases.
EEG differs from other signals due to low amplitude, high noise sensitivity, non-stationary behavior, and frequency-based interpretation.
2(e) Signal averaging in EP estimation
Signal averaging improves EP estimation by repeating stimuli and averaging responses.
Noise cancels out while the evoked response strengthens, improving signal-to-noise ratio and diagnostic accuracy.
SECTION C
3(a) ECG signal acquisition & signal quality improvement
ECG acquisition involves electrodes placed on the body, amplification, filtering, and digitization.
Signal quality is improved using proper electrode placement, shielding, filtering (baseline wander and power line noise removal), and motion artifact reduction.
3(b) Electro-Retinography (ERG) & its significance
ERG records retinal electrical activity in response to light.
It helps diagnose retinal disorders like retinitis pigmentosa, diabetic retinopathy, and macular degeneration by assessing rod and cone function.
4(a) Baseline wander & power line interference in ECG
Baseline wander is caused by respiration and body movement, while power line interference arises from electrical sources (50/60 Hz).
These artifacts distort ECG signals, affecting waveform interpretation and diagnosis accuracy.
4(b) Arrhythmia and ECG-based diagnosis
Arrhythmia is an abnormal heart rhythm.
ECG detects arrhythmias by analyzing heart rate, rhythm irregularities, QRS morphology, and timing variations.
5(a) AZTEC algorithm in biomedical signals
AZTEC adapts to ECG, EEG, and EMG by approximating signals using amplitude-zone time epochs.
It reduces data while preserving clinically important features.
5(b) Fan algorithm in biomedical signals
The Fan algorithm approximates signals using linear segments.
It handles noise by controlling error tolerance and preserves diagnostic information during compression.
6(a) Sleep-wake transition dynamics in EEG
During sleep onset, EEG shifts from alpha waves to theta and delta waves.
During wakefulness, beta and alpha activity increases, indicating alertness.
6(b) Periodogram, MEM & AR methods in EEG spectral estimation
Periodogram: Estimates power spectrum using Fourier transform
MEM: Provides high-resolution spectral estimation
AR method: Models EEG as an autoregressive process for smooth spectral estimates
7(a) LMS adaptive filter in EP estimation
The LMS filter minimizes mean square error by iteratively updating filter coefficients based on error signals.
It improves EP estimation by adapting to noise characteristics.
7(b) Wavelet detection in EP analysis
Wavelet analysis provides time-frequency representation.
Advantages include better noise handling, detection of transient signals, and improved resolution compared to traditional methods.
