Projects

Hybrid deep learning framework for gallbladder cancer classification, achieving 99.37% accuracy using quad-fold parallel CNNs and LSTM.

MATLAB-based retinal fundus segmentation using the Principal Curvature (Hessian) method with CLAHE, achieving 96.17% accuracy for early detection of diabetic retinopathy and glaucoma.

Segmentation of brain tumors from grayscale MRI images using morphological operations in MATLAB, as part of a Biomedical Image Processing lab project.

ESP8266 NodeMCU system monitoring pilot vitals and aircraft state via Blynk IoT, using MAX30100, DS18B20, and BMP180 sensors to enhance flight safety.

Wearable smart band tracking heart rate, temperature, and SpO₂ with a real-time mobile app and email alert mechanism for health irregularities.

RFID-enabled automated vending machine for menstrual hygiene products, built at FabLab KUET. 2nd Runner-Up, Intra-University Innovation Competition 2022.

Portable Arduino + TCS3200 colorimetric analyzer for urine analyte detection, with Bluetooth transmission to a companion Android app. Supports dedicated sensor and camera modes.

Secure Android telemedicine app with encrypted video consultations, medical image and prescription sharing, and real-time chat built with Java, Firebase, and ZEGOCLOUD.

Arduino-based autonomous line-following robot with IR sensors, designed and built for inter-university competition purposes.