Segmentation of Brain Tumors and Patient Survival Prediction: Methods for the BraTS 2018 Challenge

Brain tumor localization and segmentation is an important step in the treatment of brain tumor patients. It is the base for later clinical steps, e.g., a possible resection of the tumor. Hence, an automatic segmentation algorithm would be preferable, as it does not suffer from inter-rater variability. On top, results could be available immediately after the brain imaging procedure. Using this automatic tumor segmentation, it could also be possible to predict the survival of patients. The BraTS 2018 challenge consists of these two tasks: tumor segmentation in 3D-MRI images of brain tumor patients and survival prediction based on these images. For the tumor segmentation, we utilize a two-step approach: First, the tumor is located using a 3D U-net. Second, another 3D U-net – more complex, but with a smaller output size – detects subtle differences in the tumor volume, i.e., it segments the located tumor into tumor core, enhanced tumor, and peritumoral edema.

[1]  Seyed-Ahmad Ahmadi,et al.  V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation , 2016, 2016 Fourth International Conference on 3D Vision (3DV).

[2]  Klaus H. Maier-Hein,et al.  nnU-Net: Self-adapting Framework for U-Net-Based Medical Image Segmentation , 2018, Bildverarbeitung für die Medizin.

[3]  Brian B. Avants,et al.  The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS) , 2015, IEEE Transactions on Medical Imaging.

[4]  et al.,et al.  Identifying the Best Machine Learning Algorithms for Brain Tumor Segmentation, Progression Assessment, and Overall Survival Prediction in the BRATS Challenge , 2018, ArXiv.

[5]  Klaus H. Maier-Hein,et al.  No New-Net , 2018, 1809.10483.

[6]  Sébastien Ourselin,et al.  Automatic Brain Tumor Segmentation Using Cascaded Anisotropic Convolutional Neural Networks , 2017, BrainLes@MICCAI.

[7]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[8]  Thomas Brox,et al.  3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation , 2016, MICCAI.

[9]  Konstantinos Kamnitsas,et al.  Ensembles of Multiple Models and Architectures for Robust Brain Tumour Segmentation , 2017, BrainLes@MICCAI.

[10]  Luca Antiga,et al.  Automatic differentiation in PyTorch , 2017 .

[11]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[12]  Andrea Vedaldi,et al.  Instance Normalization: The Missing Ingredient for Fast Stylization , 2016, ArXiv.

[13]  Christos Davatzikos,et al.  Advancing The Cancer Genome Atlas glioma MRI collections with expert segmentation labels and radiomic features , 2017, Scientific Data.

[14]  Klaus H. Maier-Hein,et al.  Brain Tumor Segmentation and Radiomics Survival Prediction: Contribution to the BRATS 2017 Challenge , 2017, BrainLes@MICCAI.

[15]  Adria Casamitjana,et al.  Cascaded V-Net using ROI masks for brain tumor segmentation , 2018 .