Guillaume Jaume

I am a postdoctoral researcher at Harvard Medical School and Brigham and Women’s Hospital in the Mahmood Lab (since May 2022). I hold a PhD in electrical engineering and computer science from the Ecole Polytechnique Federale de Lausanne (EPFL), obtained in collaboration with IBM Research – Europe and ETH – Zurich under the supervision of Jean-Philippe Thiran and Maria Gabrani.

My research focuses on the integration of artificial intelligence (AI) tools into the clinical and research aspects of pathology with two primary objectives: (1) advancing representation learning of tissue by developing large-scale generic-purpose foundation models for histology, and (2) incorporating AI tools in drug development for better drug safety assessment, toxicity detection, and safety biomarker discovery.

News

Oct'24 HEST to appear at NeurIPS will be a Spotlight (top 2.0% of submissions).
July'24 Announcing TRACE, our new preprint on toxicity assessment and drug safety!
July'24 Announcing GEESE, our new preprint for multimodal toxicity identification!
July'24 Announcing MADELEINE to appear at ECCV, a multimodal pretraining strategy for slide representation learning. Co-authored with Anurag Vaidya.
May'24 Announcing TriPath published in Cell, the first deep learning framework for 3D pathology. Work led by Andrew H. Song.
May'24 TANGLE to appear at CVPR will be an Oral (top 0.7% of submissions.)
May'24 Announcing MMP to appear at ICML, a multimodal model for survival outcome prediction. Work led by Andrew Song and Richard Chen.
April'24 New Nature Medicine study on fairness in computational pathology. Work led by Anurag Vaidya, Richard Chen and Drew F.K. Williamson.
March'24 Announcing UNI published in Nature Medicine, a generic-purpose foundation model for histology. Work led by Richard Chen.
March'24 Announcing CONCH published in Nature Medicine, a visual-language model for histology. Work led by Ming Y. Lu.
Feb'24 Announcing TANGLE to appear at CVPR, a multimodal pretraining strategy for slide representation learning. Co-authored with Lukas Oldenburg.
Feb'24 Announcing SurvPath to appear at CVPR, multimodal model for survival outcome prediction. Co-authored with Anurag Vaidya.
Feb'24 Announcing PANTHER to appear at CVPR, a model for survival prediction. Work led by Andrew H. Song and Richard Chen.
June’23 New Nature Reviews Bioengineering article on AI for digital and computational Pathology. Co-led with Andrew H. Song.