Haifan Gong received his PhD in Computer Information Engineering from The Chinese University of Hong Kong, Shenzhen in 2025, where his research focused on developing robust and generalizable machine learning methods for biomedical image analysis. During his doctoral studies, he worked on domain generalization, structure-aware representation learning, and learning from heterogeneous supervision, with contributions spanning medical image segmentation, landmark detection, and multimodal AI. His recent work investigated how AI agents can integrate specialized models, biomedical evidence, and human feedback to support auditable clinical diagnosis, report generation, and therapeutic discovery.
Haifan’s broader research interests center on cross-scale biomedical AI, connecting molecular mechanisms, anatomical structures, and clinical observations. His work brings together robust representation learning, multimodal modeling, and agentic AI, with the goal of integrating heterogeneous evidence across biological scales to support biomedical understanding and discovery. His research is particularly motivated by the need to bridge molecular and clinical perspectives, where AI systems must connect information from molecules, tissues, organs, and patients while accounting for differences in data modalities, resolution, and uncertainty.
By integrating advances in machine learning, biomedical modeling, and human-AI collaboration, Haifan aims to develop AI systems that can reason across biomedical scales and connect biological insights with clinical evidence. His long-term goal is to advance reliable cross-scale biomedical AI that links therapeutic discovery with disease characterization and clinical decision-making, enabling discoveries at one scale to inform questions and interventions at another.