This repository contains the pretraining code for the Pillar-0 model.
A Python-based engine for processing radiology reports using the Qwen3 model with sglang for efficient batch inference. Includes quality control file generation and performance evaluation tools for comprehensive validation workflows with debug mode for faster iteration.
A comprehensive evaluation pipeline for Vision-Language Models on medical imaging tasks, with built-in support for multi-GPU processing, real-time progress tracking, and disease finding classification.
Finetuning framework for Pillar medical imaging models.
A high-performance medical image processing engine for converting DICOM and NIfTI files into ML-ready formats with intelligent windowing and compression.
monitor.ai: Non-intrusive monitoring for FDA-approved medical AI, helping healthcare organizations ensure safety and compliance without modifying validated models.
MedRAG is a multi-modal medical retrieval and generation system that combines research literature and radiological images to deliver evidence-grounded, context-aware medical insights. The system integrates semantic search, citation-aware ranking, and controlled language generation to support exploratory medical research and decision support.
The training, eval and layout for my Thesis project + paper ChestGPT.
Add a description, image, and links to the radiology-ai topic page so that developers can more easily learn about it.
To associate your repository with the radiology-ai topic, visit your repo's landing page and select "manage topics."