Yiming Ma is a postdoctoral researcher working on artificial intelligence for prostate cancer active surveillance, with a focus on patient-centred, longitudinal clinical decision support.

His research lies at the intersection of medical imaging, multimodal machine learning, and precision medicine, addressing challenges in patient selection, risk stratification, and disease progression assessment. His current work focuses on integrating imaging, clinical, demographic, and textual data into a unified Multi-Modal Fingerprint to enable personalised surveillance pathways. He has a particular interest in patient-level modelling, longitudinal disease representation, and the translation of AI methods into real-world clinical workflows.

He obtained his PhD in Mathematics of Systems at University of Warwick, with his research background rooted in computer vision and deep learning. During his doctoral research, he proposed efficient and scalable methods for crowd counting, including multiscale feature fusion, zero-inflated probabilistic modelling, and vision–language approaches based on CLIP. He also developed a robust multiview, multimodal driver monitoring system using masked multi-head self-attention, addressing real-world challenges such as sensor heterogeneity and partial observability.

Building on this methodological foundation, his current research translates advanced representation learning and multimodal fusion techniques into medical imaging and clinical AI, with an emphasis on reliability, generalisation, and deployment in constrained healthcare environments.