Medical images, such as computed tomography (CT), magnetic resonance imaging (MRI), and ultrasound, play a crucial role in providing patient-specific diagnosis, prognosis, and treatment. The value of medical imaging can be greatly amplified by robust and accurate AI-driven extraction of information from medical images. The MIA group is dedicated to developing AI methods that improve various aspects of the imaging pipeline, ranging from solving inverse problems and reconstruction to predicting complex hemodynamics-based biomarkers. To overcome challenges associated with data sparsity in medical imaging, we leverage the inherent structure within our data. By exploiting symmetries and incorporating prior knowledge, we can extract more meaningful insights even with limited data availability. Moreover, we aim to connect the appropriate data representation to each problem by learning on images, meshes, graphs, point clouds and continuous data representations. The MIA group collaborates closely with clinical partners and industry and is part of national and international externally funded research consortia with applications in prostate cancer and cardiovascular diseases. With our partners, we aim to bring artificial intelligence in precision medicine one step further.