Research in Progress | Rising Star Seminar

Modern pathology increasingly relies on multimodal data, from whole-slide images and spatial transcriptomics to clinical metadata, yet current computational tools remain limited in their ability to integrate these signals and collaborate effectively with human experts.

In this talk, Tianyu Lian will introduce two complementary systems, spEMO and TeamPath, that together move us toward a robust AI Co-Pathologist capable of assisting and augmenting expert decision-making. spEMO integrates multi-modal information especially from image sides and biomolecular sides to help clinical-decision making as well as biomarker discoveries, while TeamPath leverages the advantages of reinforcement learning to invoke the reasoning capacities of AI pathologists to make reliable outcomes. Both systems work as strong teammates in real-word challenges such as medical report generation as well as disease diagnosis, validated by physicians from Yale School of Medicine.

Tianyu Liu is a PhD candidate from Yale CBB advised by Prof. Hongyu Zhao. He works on on artificial intelligence (AI) and machine learning (ML) methods for scientific research. Having good understanding for both machine learning concepts and biological knowledge, Liu has led multiple research projects based on multimodal AI Agents and large reasoning models for computational biology research. He has published several papers in the high-impacted journals (Nature and Cell series) and ML conferences (NeurIPS, ICML, etc.). He also served as reviewers and organizers for key conferences in this field. His research is supported by fundings from NIH, NSF, Google, and OpenAI.