torchxrayvision

TorchXRayVision: A library of chest X-ray datasets and models. Classifiers, segmentation, and autoencoders.

Multi-Domain Balanced Sampling Improves Out-of-Distribution Generalization of Chest X-ray Pathology Prediction Models

Web App for Chest X-ray classification with TorchXRayVision DenseNet121.

Open, modular MCP server for medical image analysis agents: containerized models (MedSAM2, SAM 2.1, TotalSegmentator, lungmask, HD-BET, SynthStrip, nnU-Net, MONAI, TorchXRayVision), preset/custom guidelines, and a code-customizable viewer.

MedAssist, assistant médical 100 % local (Ollama + Open WebUI) : réponses sourcées HAS/OMS, relecture automatique, double lecture des radios thoraciques, mesures. Deux profils matériels : 32 Go avec carte graphique, ou PC bureautique 16 Go. Code MIT et guide d'installation.

智阅胸片——胸部 X 光智能辅助分析系统

A computer vision (CV) application that performs real-time classification of chest X-rays to detect pneumonia using TorchXRayVision. The model accurately distinguishes between normal and pneumonia cases, aiding healthcare professionals in early diagnosis and treatment.

An MCP-compliant server and client for chest X-ray classification and segmentation using TorchXRayVision.

Multi-label chest pathology classifier with Deep Learning (AUC 0.869). PyTorch + TorchXRayVision + ONNX. Includes notebook, trained models and architecture diagrams.

Research prototype: NICE NG12-based cancer symptom checklist paired with an open-source chest X-ray model (TorchXRayVision). Not a medical device.

Agentic Diagnostic Decision Support system for chest X-ray analysis — multi-agent pipeline with vision AI, PubMed RAG, and Gemini-powered clinical reasoning. Research/education only.

Local chest X-ray analysis API with Grad-CAM heatmaps