Improving Performance, Robustness, and Fairness of Radiographic AI Models with Finely-Controllable Synthetic Data

Segmentation with the Merlin backbone

an educational platform for scientific articles

[ACL-2025] CheXalign: Preference fine-tuning in chest X-ray interpretation models without human feedback

Official repository for the paper "Towards Interpretable Counterfactual Generation via Multimodal Autoregression"

[npj Digital Medicine] A generalizable 3D framework and model for self-supervised learning in medical imaging

manuscript alert system

Toward Expert-Level Medical Text Validation with Language Models

[ICML-2025] We introduce Lie group Relative position Encodings (LieRE) that goes beyond RoPE in supporting n-dimensional inputs.

Windows Subsystem for Linux

Symile is a flexible, architecture-agnostic contrastive loss that enables training modality-specific representations for any number of modalities.

Official PyTorch implementation of FlowMo.

Torch utilities for 3D medical imaging

Cloud-Based Deep Learning ☁️

Merlin is a 3D VLM for computed tomography that leverages both structured electronic health records (EHR) and unstructured radiology reports for pretraining.

[MIDL 2025] Efficient Automated Interpretation of Medical Images with Large-Scale Generalizable Autoencoders

preliminary quantification of medical imaging data