MeDSAI

Research Focus

The Medical Data Sciences and Artificial Intelligence (MeDSAI) research group is a highly multidisciplinary team dedicated to the development and application of data science and artificial intelligence methods in medical research and healthcare. Its work focuses on analysis and integration of diverse large-scale medical data types within a big data framework to advance clinical decision-making, deepen disease understanding, and enable personalized medicine.

Team and Expertise

MedSAI brings together expertise in machine learning, statistical modeling, biomedical informatics, and computer science, combined with strong domain knowledge in medicine and the life sciences. A particular emphasis is placed on the use and development of large language models (LLM) and foundation models for development of advanced multi-modal AI frameworks.

Research activities focus on both structured and unstructured data and integration of multi-modal data including electronic health records, laboratory and biomarker data, physiological signals, omics data, imaging and other digital health data derived from clinical and real-world settings.

Image-based data types and analysis are not the primary focus of the group activity.

A key objective is the development of robust, interpretable, and clinically meaningful data-driven AI models for diagnosis, prognosis, risk stratification, and therapy support. 

Current projects

The group places strong emphasis on explainable and responsible AI, translational research, and close collaboration with clinical and scientific partners to enable real-world impact in healthcare. The group’s research projects are developed in close collaboration with clinical and scientific partners and try to address real-world healthcare challenges. They include:

• Development of AI and data science methods for diagnosis, prognosis, risk stratification, and therapy support
• Research on multi-modal AI frameworks integrating clinical, laboratory, omics, physiological, imaging and digital health data
• Development and application of large language models (LLMs) and foundation models for medical applications
• Integration and analysis of structured and unstructured healthcare data within large-scale medical data frameworks
• Development of interpretable and clinically meaningful AI methods for decision support and personalized medicine
• Translational research and validation of AI solutions in real-world healthcare settings