In precision medicine, each person’s unique characteristics are taken into account to focus on individual outcomes instead of general population statistics. Artificial Intelligence (AI), particularly deep learning, has made substantial advances in predictive models for precision medicine. Currently, the two most important domains are radiology, i.e., “radiomics”, and pathology, i.e., “pathomics”. While radiomics and pathomics often have similar goals and contain complementary information, these research fields are largely separated and rarely combined.
To harness their complementary information, we develop AI for Integrated Diagnostics methods to co-learn from multiple modalities. We focus on medical imaging in oncology, but aim to develop method- and disease-agnostic to generalize our methods and make pan-cancer or even pan-disease models. Our vision is to ultimately join the forces of all relevant modalities for these diseases to gain maximum impact. To facilitate adaptation in real-world clinical practice, we develop trustworthy AI and intensively collaborate with clinicians from various disciplines.
