A PyTorch implementation of a diffusion-model + U-Net pipeline for automated multi-class segmentation of knee MRI scans. Includes dataset utilities, training/evaluation workflows, and 2D/3D inference scripts, with Docker and conda support for reproducibility.
computer-vision deep-learning pytorch medical-imaging unet mri-segmentation diffusion-models semant healthcare-ai knee-mri
-
Updated
Sep 6, 2025 - Python