Serena Yeung

Serena Yeung is an Assistant Professor of Biomedical Data Science at Stanford University School of Medicine, with courtesy appointments in Computer Science and Electrical Engineering. She leads the Medical AI and Computer Vision Lab (MARVL) at Stanford and serves as Associate Director of Data Science for the Stanford Center for Artificial Intelligence in Medicine & Imaging (AIMI). She is also affiliated with the Stanford Clinical Excellence Research Center (CERC).

Dr. Yeung's research focuses on computer vision, machine learning, and deep learning with applications to healthcare. Her work spans surgical video analysis, medical imaging AI, human motion analysis, and vision-language models for clinical applications. She has pioneered approaches for analyzing patient activities in ICU settings, surgical skill assessment, and hand hygiene monitoring in hospitals using computer vision systems.

Her recent publications reveal a strong trend toward multimodal AI systems applied to medical imaging, with increasing focus on foundation models, surgical AI, medical vision-language systems, and 3D human motion analysis for clinical applications. Her work bridges computer vision fundamentals with practical healthcare implementations.

• Best Paper Award at NIPS 2017 Machine Learning for Health Workshop

Dr. Yeung leads significant research initiatives including the Medical AI and Computer Vision Lab (MARVL) and serves as Associate Director of Data Science for the Stanford Center for Artificial Intelligence in Medicine & Imaging (AIMI). Her research has been supported by collaborations with major healthcare institutions and technology companies, focusing on translating computer vision research into clinical practice.

Dr. Yeung directs the Medical AI and Computer Vision Lab (MARVL), which focuses on developing computer vision systems for healthcare applications. The lab's work spans surgical video analysis, ICU patient monitoring, medical imaging, and clinical decision support systems using state-of-the-art computer vision and deep learning techniques.