M.Sc (Engg.) in Information & Communication Technology – Ongoing
Comilla University

RESEARCH EXPERIENCE

An Unified Quantum-Classical Model for Noisy Label Medical Image Binary Classification
Developed an 8-qubit quantum-classical model combining CNN and QNN for robust classification on OrganMNIST and PneumoniaMNIST under symmetric label noise (10–30%).
Supervisor: Amena Begum, Assistant professor, Department of ICT, Comilla University
AI-Powered Inventory Management Using YOLOv8 Object Detection and Integrated OCR
for Package Label Recognition
Built a warehouse automation tool using YOLOv8 and Tesseract, achieving 0.862 mAP50 and 92.6% OCR accuracy on a Daraz-based dataset. Enabled tracking, anomaly detection, and multilingual label reading.
Supervisor: Alimul Rajee, Assistant professor, Department of ICT, Comilla University