RSNA-MICCAI Brain Tumor Segmentation (BraTS) Challenge 2021

Spyridon Bakas, Christopher Carr, Adam E. Flanders, Jayashree Kalpathy–Cramer, John Mongan, Bjoern Menze, Luciano M. PrevedelloZenodo 2021 · DOI

Abstract

The Brain Tumor Segmentation (BraTS) 2021 challenge is the 10th anniversary of BraTS and this year is jointly<br> organized by RSNA and MICCAI. Compared to the previous BraTS instances, the RSNA MICCAI BraTS challenge<br> utilizes multi-institutional multi-parametric magnetic resonance imaging (mpMRI) scans from both pre-operative<br> and post-operative timepoints, i.e., including surgically imposed cavities (instead of only pre-operative scans), and<br> focuses only on glioblastoma (instead of all glioma). Glioblastoma, and diffuse astrocytic glioma with molecular features of glioblastoma (WHO Grade 4 astrocytoma),<br> are the most common and aggressive malignant primary tumor of the central nervous system in adults, with<br> extreme intrinsic heterogeneity in appearance, shape, and histology. Glioblastoma patients have very poor<br> prognosis, and the current standard of care comprises surgical resection followed by radiotherapy and<br> chemotherapy. MGMT (O[6]-methylguanine-DNA methyltransferase) is a DNA repair enzyme that the methylation<br> of its promoter in newly diagnosed glioblastoma has been identified as a favorable prognostic factor and a<br> predictor of chemotherapy response. Thus determination of MGMT promoter methylation status in newly<br> diagnosed glioblastoma can influence treatment decision making. In consistency with the previous BraTS instances, BraTS'21 focuses on (Task 1) the evaluation of state-of-the-art<br> methods for the segmentation of intrinsically heterogeneous brain glioblastoma sub-regions in mpMRI scans.<br> Furthemore, to pinpoint the clinical relevance of this segmentation task, BraTS’21 also focuses on (Task 2) the<br> evaluation of methods to predict the MGMT promoter methylation status at the pre-operative scans, via<br> integrative analyses of quantitative imaging phenomic features and machine learning algorithms.<br> Ground truth annotations of the tumor sub-regions are created and approved by expert neuroradiologists for<br> every subject included in the training, validation, and testing datasets to quantitatively evaluate the predicted<br> tumor segmentations of Task 1, whereas the quantitative evaluation of Task 2 is performed according to clinical<br> data. Participants are free to choose whether they want to focus only on one or both tasks.
BibTeX
@inproceedings{bakas2021rsna,
  author = {Spyridon Bakas and Christopher Carr and Adam E. Flanders and Jayashree Kalpathy–Cramer and John Mongan and Bjoern Menze and Luciano M. Prevedello},
  title = {{RSNA-MICCAI Brain Tumor Segmentation (BraTS) Challenge 2021}},
  booktitle = {Zenodo},
  year = {2021},
  doi = {10.5281/zenodo.4575162},
}

Loading related content...

Figure