Abstract
Fluid-attenuated inversion recovery (FLAIR) magnetic resonance imaging (MRI) scans are important for diagnosis, treatment planning, and monitoring of various brain tumors. Depending on the tumor type, the FLAIR hyperintensity volume is an important measure to assess the tumor volume, surrounding vasogenic edema, or treatment induced changes, such as gliosis. Automatic segmentation would therefore be valuable in the clinic and in clinical trials. In this study, around 5000 FLAIR images of various brain tumors types and acquisition time points, from different neurosurgical centers, were used to train a unified FLAIR hyperintensity segmentation model using an Attention U-Net architecture. The performance was compared against dataset-specific models and was validated on different tumor types, acquisition time points, and against BraTS. The unified model achieved an average Dice score of 88.65% for pre-operative meningiomas, 80.08% for pre-operative metastases, 90.92% for pre-operative and 84.60% for post-operative gliomas from BraTS, and 84.47% for pre-operative and 61.27% for post-operative lower grade gliomas. In addition, the results showed that the unified model achieved comparable segmentation performance to the dataset-specific models on their respective datasets. The documented generalization across tumor types and acquisition time points is a strong indicator for efficient deployment in a clinical setting. The model has been integrated into Raidionics, an open-source software for CNS tumor analysis.