Learn-Morph-Infer: A new way of solving the inverse problem for brain tumor modeling

Ivan Ezhov, Kevin Scibilia, Katharina Franitza, Felix Steinbauer, Suprosanna Shit, Lucas Zimmer, Jana Lipková, Florian Kofler, Johannes C. Paetzold, Luca Canalini, Diana Waldmannstetter, Martin J. Menten, Marie Metz, Benedikt Wiestler, Bjoern H. Menze

Abstract: Highlights•Treatment of brain tumors benefits by accessing the spatial distribution of tumor cells.•Tumor simulations complement imaging information by providing estimates of cell distribution.•We introduce a learnable methodology for inferring the patient-specific tumor distribution.•The method achieves real-time performance on widely available hardware.