MedAI

Our mission is to shape the future of healthcare through innovative AI solutions, aiming to enhance personalised medicine and advance equitable AI across fields such as cardiology, digital pathology, and ophthalmology.

The MedAI Lab is based in the School of Computer Science at the University of Birmingham. We closely collaborate with members of College of Medicine and Health, local hospitals, and international partners to ensure our research delivers real-world impact.

Our research spans multiple domains of medical AI, focusing on developing robust, interpretable, and clinically-relevant AI systems. Key research interests are:

• Digital health and digital twins, e.g., 3D reconstruction, multi-view geometry
• Generative AI for medical imaging, e.g., image and video synthesis
• Transfer learning, e.g., domain adaptation, few-/zero-shot learning
• Representation learning with limited supervision, e.g., self-/semi-/weakly-supervised methods
• Multimodal medical AI, e.g., vision-language models, LLMs, and clinical AI agents
• Foundation models in healthcare, with a focus on adaptation and evaluation