About this Research Topic
Gliomas are the most common primary brain malignancies, with different degrees of aggressiveness, variable prognosis and various heterogeneous histologic sub-regions, e.g., necrotic, enhancing and non-enhancing tumor core, as well as peritumoral edematous and invaded tissue. This intrinsic heterogeneity is also portrayed in their radiographic imaging phenotype (appearance and shape), as their sub-regions are described by varying intensity profiles disseminated across multimodal MRI (mMRI) scans, reflecting varying tumor biological properties. Due to their highly heterogeneous phenotype, the accurate identification, segmentation, and quantification of related tumor sub-structures in mMRI, is one of the most challenging tasks in medical image analysis.
To foster research in this field, the “Multimodal Brain Tumor Segmentation” (BraTS) benchmark has been focusing since 2012 on creating a publicly-available large-scale multi-institutional dataset with pre- and post-operative mMRI of low- and high-grade glioma patients. Since 2012, BraTS has offered over 1000 harmonized, pre-processed, curated, and expert-annotated mMRI brain scans, and has been used in the development and evaluation of new brain and tumor image analysis methodologies. Beyond its multiclass segmentation task on mMRI, also offers tumor grading, multi-temporal change, and radiomic analyses tasks, such as survival prediction. Furthermore, BraTS makes use of the NIH open resource, namely ‘The Cancer Imaging Archive’ (TCIA), thereby facilitating multi-disciplinary radiogenomic research by linking to the corresponding genetic information publicly available in ‘The Cancer Genome Atlas’ (TGCA) of NIH. This benchmark challenge, organized in conjunction with the MICCAI conference, has attracted several hundred contributions from research groups worldwide, spanning across the fields of medical image analysis, computer vision, machine learning, and radiology, towards advancing image computing technologies and promoting the BraTS challenge as one of the reference biomedical image computing benchmarks.
This Research Topic invites contributions spanning across the fields of medical image analysis, machine learning, and radiology, towards advancing the following topics and while using data from the BraTS challenge, or the TCIA collections of TCGA-GBM, TCGA-LGG, and IvyGAP, as well as their corresponding genomic data from TCGA. (Additional use of private/public data is welcome.)
1) Brain Tumor Segmentation (incl. high-performing BraTS participants),
2) Radiomic predictions tasks, e.g., survival prediction (incl. high-performing BraTS participants),
3) Radiogenomic analysis,
4) Brain tumor image analysis advancements beyond segmentation, e.g., normalization, registration, skull-stripping,
5) Systematic performance evaluation of related existing approaches/tools,
6) Descriptions and analysis of resources that have been made available for BraTS/TCIA, or using BraTS/TCIA data, and other challenge outcomes,
7) Analysis results of the BraTS competition,
8) Studies comparing the BraTS challenge with other brain lesion segmentation challenges,
9) Studies comparing the BraTS data (or results from using these) with other private/public datasets,
10) Demonstration of the impact of the BraTS challenge in different communities.
To foster research in this field, the “Multimodal Brain Tumor Segmentation” (BraTS) benchmark has been focusing since 2012 on creating a publicly-available large-scale multi-institutional dataset with pre- and post-operative mMRI of low- and high-grade glioma patients. Since 2012, BraTS has offered over 1000 harmonized, pre-processed, curated, and expert-annotated mMRI brain scans, and has been used in the development and evaluation of new brain and tumor image analysis methodologies. Beyond its multiclass segmentation task on mMRI, also offers tumor grading, multi-temporal change, and radiomic analyses tasks, such as survival prediction. Furthermore, BraTS makes use of the NIH open resource, namely ‘The Cancer Imaging Archive’ (TCIA), thereby facilitating multi-disciplinary radiogenomic research by linking to the corresponding genetic information publicly available in ‘The Cancer Genome Atlas’ (TGCA) of NIH. This benchmark challenge, organized in conjunction with the MICCAI conference, has attracted several hundred contributions from research groups worldwide, spanning across the fields of medical image analysis, computer vision, machine learning, and radiology, towards advancing image computing technologies and promoting the BraTS challenge as one of the reference biomedical image computing benchmarks.
This Research Topic invites contributions spanning across the fields of medical image analysis, machine learning, and radiology, towards advancing the following topics and while using data from the BraTS challenge, or the TCIA collections of TCGA-GBM, TCGA-LGG, and IvyGAP, as well as their corresponding genomic data from TCGA. (Additional use of private/public data is welcome.)
1) Brain Tumor Segmentation (incl. high-performing BraTS participants),
2) Radiomic predictions tasks, e.g., survival prediction (incl. high-performing BraTS participants),
3) Radiogenomic analysis,
4) Brain tumor image analysis advancements beyond segmentation, e.g., normalization, registration, skull-stripping,
5) Systematic performance evaluation of related existing approaches/tools,
6) Descriptions and analysis of resources that have been made available for BraTS/TCIA, or using BraTS/TCIA data, and other challenge outcomes,
7) Analysis results of the BraTS competition,
8) Studies comparing the BraTS challenge with other brain lesion segmentation challenges,
9) Studies comparing the BraTS data (or results from using these) with other private/public datasets,
10) Demonstration of the impact of the BraTS challenge in different communities.
Keywords: Brain tumor, Image segmentation, Multiparametric MRI, Benchmark, Machine Learning
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
