Code for reproducing the experiments on large-scale pre-training and transfer learning for the paper "Effect of large-scale pre-training on full and few-shot transfer learning for natural and medical images" (https://arxiv.org/abs/2106.00116)

The best single model performance on the CheXpert chest X-ray classification competition

Ensemble based transfer learning approach for accurately classifying common thoracic diseases from Chest X-Rays

Codebase for Imperial MSc AI Group Project: How Well Do Self-Supervised Models Transfer to Medical Imaging?

Chest X-ray Classification

X-Ray diagnosis and heatmap visualization of disease area using deep learning models like VGG-16 and Grad-CAM with 86% accuracy

Deep Learning AI capable of diagnosing 14 chest diseases with 90% accuracy . Performing Image Processing and Chest XRay analysis and exports AI generated Medical Report.

This repository explores the application of Self-Supervised Learning techniques such as MoCo, SimCLR, BYOL, and SwAV, for medical image representation using the CheXpert dataset. The project aims to classify multiple thoracic diseases while minimizing reliance on labeled data and addressing challenges like class imbalance and label uncertainty.

Enhancing domain generalization in the AI-based analysis of chest radiographs with federated learning

Learning to Generalize towards Unseen Domains via a Content-Aware Style Invariant Framework for Disease Detection from Chest X-rays