Top 1 solution of Chexpert

The official implementation of "Delving into Masked Autoencoders for Multi-Label Thorax Disease Classification"

ICVGIP' 18 Oral Paper - Classification of thoracic diseases on ChestX-Ray14 dataset

Reproducing CheXpert paper using the PyTorch library.

Code for CheXpert Challenge 2019 Top 1 && Top 2 solution

[MICCAI 2023, top 14%] Official repository of paper: Distilling BlackBox to Interpretable models for Efficient Transfer Learning

Detecting Shortcuts in Medical Images - A Case Study in Chest X-rays - ISBI 2023

Securing Collaborative Medical AI by Using Differential Privacy

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

Proposed solution and baseline for CheXpert dataset, implemented in PyTorch. CheXpert is a large dataset of chest X-rays and competition for automated chest x-ray interpretation by Stanford ML Group.

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

Chest X-ray screening research with explicit distribution shift evaluation, per-disease binary models and validated Grad-CAM localization.

Projective Transformation Rectification Network Official Implementation

Code repository for training multi-label classification models on the CheXpert Chest X-ray dataset.

Chest X-ray classification

Multilabel Chest Xray Disease Diagnosis

Using Deep learning based predictive models for chest x-ray diagnosis using the CheXpert Dataset.

Chest X-ray multi-label classifier built on DenseNet-121 with channel attention and Grad-CAM explainability. Trained on CheXpert (14 pathologies) achieving 0.873 mean AUC on competition labels. Features clinical metrics (sensitivity, specificity), uncertainty label handling, and a single-image inference CLI with interpretable reports.