PI: Martin Urschler
Focus: The focus of our work is on the medical image analysis of primarily radiological data sources, such as X-ray images, Computed Tomography (CTs), or Magnetic Resonance Imaging (MRIs), but also includes the analysis of microscopic images. Modern artificial intelligence methods that use deep neural networks to predict anatomical and pathological structures are at the center of our current research projects. Methodologically, we investigate Convolutional Neural Networks (CNNs) and transformer architectures for tasks such as the robust localization of anatomical landmarks, the segmentation and precise delineation of organs and pathologies, as well as the distinction between malignant and benign tumors as well as tumor grading. We have numerous collaborations with various clinical departments from within the Medical University of Graz, as well as external collaborations, which has resulted in compelling research questions ranging from diverse regions of the body. Examples are X-ray images of bones located in the hands, the analysis of enchondromas in knee MRI, research involving the blood vessels located within the lungs, the analysis of heart chambers, and research concerning spine vertebrae in CT.
Network: Important collaborations of our research team within the Medical University Graz currently include working with the Computational Cardiology Lab at the Chair of Medical Physics and Biophysics, the Divsion of Oral Surgery and Orthodontics, the Dvision of General Radiological Diagnostics, the Dvision of Pediatric Radiology, the Department for Orthopedics and Traumatology, the Division of Cardiology, the Signaling research group at the Chair of Cell Biology, Histology and Embryology, and the Neuroimaging research group at the Department for Neurology .
Within Graz, we also collaborate with the Institute for Computer Graphics and Vision at the TU Graz, and the Institute for Mathematics and Scientific Computing at the University of Graz. International collaborations currently exist with, among others, the School of Computer Science at the University of Auckland, New Zealand, as well as with the Department of Computer Engineering at the University of Rijeka, Croatia. Our research team is a member of the international MICCAI society (Medical Image Computing and Computer-Assisted Intervention).
