Health Intelligence

The goal of our group is to build clinically ready foundation models that span the spectrum of healthcare, as well as world models that reproduce, predict, and interact with clinical observations. Ultimately, we aim to extend intelligence from virtual environments to physical clinical scenarios.

The next step for medical AI is scaling. The Bitter Lesson (Richard S. Sutton, 2019) reminds us that “general methods that leverage computation are ultimately the most effective.” Yet scaling medical AI faces intrinsic challenges: from data shortage and technical readiness to ethical concerns, financial feasibility, cultural viability, and business sustainability. Nevertheless, we believe the timing is right: breakthroughs in spatial intelligence, generative AI, and multimodal learning are opening a clear trajectory forward.

We are fortunate to be supported by growing public datasets, even larger proprietary clinical data resources, established medical collaborations, and powerful compute.

Current research topics

•    AI for healthcare
•    Generative AI and multimodal learning
•    Spatial intelligence and 3D/geometric deep learning
•    Digital twins and world models
•    Agentic AI
 

The HINT Lab is led by Assistant Professor Jiancheng Yang (Jiancheng Yang's website).

The group operates within the Department of Electrical Engineering and Automation at Aalto University School of Electrical Engineering and ELLIS Institute Finland, a research hub in AI and Machine learning.