UNI2-h is an open-source self-supervised vision transformer developed by Mahmood Lab for pathology image analysis. It generates rich feature embeddings from histology slides and supports integration via the timm library. The model is intended for non-commercial academic research in medical AI and computational pathology.
UNI2 H sits in PulseGate's Other AI category. It focuses on obtaining high-quality, generalizable embeddings from pathology images without extensive labeled data. UNI2 H is an open-source project aimed at AI researchers and pathologists. The project is open source (Open Source). It ships for the web and API.
Mahmood Lab builds and maintains UNI2 H, and it first shipped in 2024. Among its 4 catalogued features are self-supervised learning, vision transformer, and histology analysis.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do