Research Portfolio

Research Interests

• Biomedical Signal Processing 

• Machine Learning

• Generative AI

• Wearable Devices for Healthcare Monitoring

M.Sc. Engineering Thesis

Title: Automatic Detection Of Paroxysmal Atrial Fibrillation By Machine Learning Approaches Using Electrocardiogram

• Worked on automated atrial fibrillation (AF) detection using short-term ECG signals in the presence of other arrhythmia beats.

• Investigated heart rate variability (HRV), atrial activity, and Poincaré image-domain analysis.

• Analyzed the possible solutions to discriminate ectopic beats, like premature atrial contraction (PAC) and premature ventricular contraction (PVC), from AF.

• The image-domain analysis of the Poincaré trajectory of the HRV exhibited superior AF discriminating characteristics.

• Concluded that an integration of HRV dynamics with atrial activity-related information can trace AF efficiently in the presence of PAC/PVC or other arrhythmia beats.

Research Projects

• Title: Real-Time Automated Cardiovascular Tachyarrhythmia Monitoring System (Ongoing)

Funded by DRE, RUET

Selected UG Thesis Supervision 

• "SleepNet: A Hybrid Deep Learning Approach for Sleep Stage Classification Using Single Lead  Electrocardiogram" By Shoeb Khan.

• "A Deep CNN-BiLSTM Model for Detection of Atrial Fibrillation Using Short Term Electrocardiogram" By Nazmul Haque.