AI & Semantics

The analysis of extensive medical data for clinical decisions requires not only the recognition of statistical correlations, but also an understanding of causal relationships in biological processes. The use of causal and data-driven models makes them more robust and understandable. Machine learning methods build a bridge between clinical information, imaging, laboratory values and electronic health records for a holistic picture of the patient. The development of evidence synthesis methods that integrate heterogeneous data sources and are based on counterfactual methods, meta-analysis and artificial neural networks is very promising. The formation of junior research groups on the topics of AI and causality, AI and bioinformatics, AI systems in multi-omics data and statistical evidence in AI systems is intended to tap into the potential in the area of "AI and semantics".