Deep learning automates bidimensional and volumetric tumor burden measurement from MRI in pre- and post-operative glioblastoma patients

Jakub Nalepa, Krzysztof Kotowski, Bartosz Machura, Szymon Adamski, Oskar Bozek, Bartosz Eksner, Bartosz Kokoszka, Tomasz Pekala, Mateusz Radom, Marek Strzelczak, Lukasz Zarudzki, Agata Krason, Filippo Arcadu, Jean Tessier

Abstract: Highlights•We propose the algorithm for segmenting pre- and post-operative gliomas from MRI.•We fully automate the volumetric and bidimensional glioma measurements.•We introduce a comprehensive and rigorous approach to generate ground-truth data.•We perform a thorough experimental study over large and heterogeneous MRI dataset.•The algorithm matches or exceeds performance of radiologists in a fraction of time.