Related Experiment Video
Updated: Jul 27, 2025
08:41
Patient-Specific Polyvinyl Alcohol Phantom Fabrication with Ultrasound and X-Ray Contrast for Brain Tumor Surgery Planning
Published on: July 14, 2020
8.5K
The Brain Tumor Segmentation (BraTS) Challenge 2023:
Anahita Fathi Kazerooni1,2,3, Nastaran Khalili1, Xinyang Liu4
1Center for Data-Driven Discovery in Biomedicine (D3b), Children's Hospital of Philadelphia, Philadelphia, PA, USA.
Arxiv
|June 9, 2023
Summary
The CBTN-CONNECT-DIPGR-ASNR-MICCAI BraTS-PEDs 2023 challenge benchmarks AI for segmenting pediatric brain gliomas. This aims to accelerate diagnosis and improve treatment for children with these rare, aggressive tumors.
Area of Science:
- Neuro-oncology
- Artificial Intelligence
- Medical Imaging
Background:
- Pediatric central nervous system tumors are the leading cause of cancer-related death in children, with high-grade gliomas having a poor prognosis.
- Rarity of pediatric brain tumors leads to diagnostic delays and reliance on outdated treatment strategies, hindering clinical trial progress.
Purpose of the Study:
- To introduce the first Brain Tumor Segmentation (BraTS) challenge focused on pediatric brain tumors.
- To benchmark volumetric segmentation algorithms for pediatric brain gliomas using standardized metrics.
Main Methods:
- Utilizing multi-parametric structural MRI (mpMRI) data from international pediatric neuro-oncology consortia.
- Evaluating AI models trained on BraTS-PEDs data using validation and unseen test datasets of high-grade pediatric gliomas.
Main Results:
- The challenge facilitates the development of automated segmentation techniques for pediatric brain gliomas.
- Standardized quantitative performance evaluation metrics are employed across the BraTS 2023 cluster of challenges.
Conclusions:
- The BraTS-PEDs 2023 challenge fosters collaboration between clinicians and AI/imaging scientists.
- Accelerated development of automated segmentation can significantly benefit clinical trials and improve care for pediatric brain tumor patients.
