medical-imaging-with-deep-learning

The source code for the Layer Ensembles paper published in MICCAI 2022 (Singapore).

[MIDL 2022 Oral] Learning Morphological Feature Perturbations for Calibrated Semi Supervised Segmentation

NeuroCure is a cutting-edge project focused on the detection and classification of brain tumors, leveraging the power of deep learning for advanced medical image analysis. Developed using TensorFlow and a variety of custom models, this initiative aims to deliver accurate and efficient identification of brain tumors from MRI scans.

Officail Pytorch implementation for: Class Attention to Regions of Lesion for Classification on Imbalanced Data. (MIDL-2019)

Leukemia cell detection system using advanced segmentation and a fine-tuned ResNet50 model, with a complete backend API and Streamlit-based frontend.

Official Repository for publication 'The Invisible Gorilla Effect in Out-of-distribution Detection'

Official implementation of the published paper "Tri-Net: Unified Deep Learning for Skin Lesion and Symptom-Based Monkeypox Detection" (Scopus Q1, SCI-E, IF 3.9), featuring a reproducible Tri-Net v2 framework with modern backbones, feature fusion, explainability, and benchmarking.

Brain tumor classification from MRI scans using CNN benchmarking (Xception, EfficientNetB4, ResNet-50) — Presented at IC3SE 2025 .

This repository is a curated collection of evolving machine learning projects—from bite-sized real-world use cases like diabetes prediction to more advanced pipelines integrating MLOps workflows. Every weekly build is crafted to deepen understanding, spark creative experimentation, and push the boundaries of applied AI.

3D Brain Tumor Segmentation using U-Net and MONAI on the BraTS 2020 dataset to optimize segmentation performance across multiple MRI modalities.

Enhanced U²-Net pipeline for generalized retinal vessel segmentation across DRIVE, HRF, CHASE_DB1, and STARE with patch-based training, hybrid enhancement, and TTA evaluation.

Production-grade medical AI framework using BioMedCLIP (ViT-B/16) for multi-label chest X-ray pathology detection. Features Monte Carlo Dropout for uncertainty quantification, per-class threshold calibration, ViT-GradCAM explainability, and LLaMA-3.3 reporting. FastAPI + Next.js.

Production-ready deep learning system for automated detection and classification of eye diseases using 10 ML models and 4,217 balanced retinal images.

Hybrid Affine–Residual Deep Learning Framework for Tumor-Aware Multimodal Brain MRI Registration with NSGA-Based Loss Optimization.

PyTorch code for Efficient Vision Mamba MRI super-resolution, Medical Physics 2026.

SCC-ALAR: An experimental study on skin lesion classification using hybrid CNN-Transformer architectures. This project explores the impact of architectural complexity versus inductive bias on the imbalanced HAM10000 dataset, featuring custom weighting schemes and pre-extracted feature embeddings.

Multi-class brain tumor segmentation from MRI using 3D Attention U-Net

Predicting Pregnancy Status by Multimodal ML Pipeline Using the Dataset for Fetus Framework

A deep learning system that analyzes medical images (X-rays, CT scans, MRIs) to assist healthcare professionals in detecting diseases like cancer, pneumonia, and fractures. Implements multi-modal learning with explainable AI for clinical trustworthiness.

Hybrid ML pipeline using FCM segmentation, EfficientNetB4 feature extraction, and XGBoost for leukemia detection.