MIT Releases FleXray, an Open X-ray Foundation Model Segmenting 60 Anatomical Structures
rishabh16_ · x · 2026-09-24
Researchers from MIT CSAIL, MGH, and Harvard Medical School (Victor Ion Butoi et al.) released FleXray, a universal foundation model for clinical X-ray anatomy segmentation.
- Trained on physics-based CT-to-DRR projections, real radiographs, and generative-model-enhanced DRRs.
- Predicts 60 anatomical structures across body regions, views, and acquisition settings, outperforming existing generalist baselines on diverse real X-ray datasets.
- Fully open: paper, code, weights on Hugging Face, training data, tutorial, plus an MCP connector.
- Browser demo runs entirely on-device via WebGPU/WebAssembly; ensemble of five models with 16-sample test-time augmentation.
- Explicitly a research demo, not a medical device.
More from Research
- Ross Girshick posts CVPR slides recounting 1.5 decades of object detection and its link to the LLM era — CatAstro_Piyush · 2026-11-11
- KAIST's WikiTrace: evidence-linked memory fixes agent "memory-induced degradation" — dair_ai · 2026-11-11
- UHN researchers unveil DynaMELD, an ML algorithm to prioritize liver transplant waitlists more equitably — VectorInst · 2026-11-11
- What it takes to squeeze performance: EMA, cosine scheduling and custom dead-cluster revival — chrisemoody · 2026-11-11
- Trainable JWST model boosts exoplanet sensitivity 2–15x in Nature paper — bravo_abad · 2026-11-11
- Ben Todd's alt log-log plot shows task horizons hitting a singularity in under 12 months — ben_j_todd · 2026-11-11