Medical imaging processing for AI applications.

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

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

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.

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

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

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.

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.

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