Hounsfield LabCurated by a practicing neuroradiologist
Open-source medical AI, measured like a lab value.
A clinician-curated directory of downloadable models, toolkits and infrastructure for building medical AI, with license, update and maintenance status at a glance.
Featured: Pillar-0
An open radiology foundation model from YalaLab. Pretrained on 42,990 abdomen-pelvis CTs, 86,411 chest CTs, 14,348 head CTs and 11,543 breast MRIs, with RATE, an LLM-based framework that extracts labels for 366 radiologic findings from reports. The authors release code, weights and evaluation tools.
Project page · Paper · Weights
LICECL-2.0UPD2026-10-12STATUSMaintainedCOMMPermissive