The Tissue Image Analytics (TIA) Centre hosts a seminar series for those interested in computational pathology. The series invites researchers and leaders in computational pathology and related fields to present their work and foster thought-provoking discussions.

This seminar features Dr Duan Zhiyun from Fudan University, China.

Abstract

Foundation models are rapidly reshaping biomedical research and hold great promise for oncology. Yet despite impressive benchmark performance, their impact in real clinical settings remains limited. A key reason is that most current models are designed as narrow specialists optimized for static tasks, while clinical oncology requires more generalist systems. Effective clinical support demands models that can integrate multimodal information, follow the longitudinal course of disease, and incorporate patient perspectives into decision-making. Rethinking model design along these dimensions will be crucial if foundation models are to move beyond benchmarks and become trusted partners in cancer care.

Bio

Duan is a medical doctor at Zhongshan Hospital of Fudan University, Shanghai, China. He has published as first author in several great journals in medicine, including Cell Reports Medicine, Nature Cancer, and Journal of Translational Medicine over the past three years.