Analysis of the BraTS 2023 Intracranial Meningioma Segmentation Challenge

2024-05-16

Machine Learning for Biomedical Imaging

AI Summary

This study addresses the challenge of automatic MRI segmentation of intracranial meningiomas—benign, multifocal, and morphologically heterogeneous tumors. To this end, we organized the first international multicenter challenge and constructed the largest publicly available expert-annotated dataset to date, comprising multi-sequence (T2, FLAIR, T1, T1-Gd) scans and multi-compartment labels (enhancing tumor, non-enhancing tumor, FLAIR hyperintense region). We established a systematic multicenter annotation benchmark and introduced a lesion-level evaluation framework combining Dice Similarity Coefficient (DSC) and 95% Hausdorff Distance (HD), revealing skull-edge artifacts as a critical performance bottleneck and motivating standardized preprocessing. Our method employs a deep learning architecture with multi-sequence input, joint multi-label training, and skull-stripping augmentation. The winning model achieved lesion-level DSCs of 0.976 (enhancing tumor), 0.976 (tumor core), and 0.964 (whole tumor), with mean DSC > 0.87 across compartments—setting the new state-of-the-art benchmark for automatic meningioma segmentation.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Intelligent Robots: Multi-Robot Systems

Application Category

• Web Mining and Content Analysis: Machine learning and data science for the Web
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

We describe the design and results from the BraTS 2023 Intracranial Meningioma Segmentation Challenge. The BraTS Meningioma Challenge differed from prior BraTS Glioma challenges in that it focused on meningiomas, which are typically benign extra-axial tumors with diverse radiologic and anatomical presentation and a propensity for multiplicity. Nine participating teams each developed deep-learning automated segmentation models using image data from the largest multi-institutional systematically expert annotated multilabel multi-sequence meningioma MRI dataset to date, which included 1000 training set cases, 141 validation set cases, and 283 hidden test set cases. Each case included T2, FLAIR, T1, and T1Gd brain MRI sequences with associated tumor compartment labels delineating enhancing tumor, non-enhancing tumor, and surrounding non-enhancing FLAIR hyperintensity. Participant automated segmentation models were evaluated and ranked based on a scoring system evaluating lesion-wise metrics including dice similarity coefficient (DSC) and 95% Hausdorff Distance. The top ranked team had a lesion-wise median dice similarity coefficient (DSC) of 0.976, 0.976, and 0.964 for enhancing tumor, tumor core, and whole tumor, respectively and a corresponding average DSC of 0.899, 0.904, and 0.871, respectively. These results serve as state-of-the-art benchmarks for future pre-operative meningioma automated segmentation algorithms. Additionally, we found that 1286 of 1424 cases (90.3%) had at least 1 compartment voxel abutting the edge of the skull-stripped image edge, which requires further investigation into optimal pre-processing face anonymization steps.

Problem

• Developed deep-learning models for meningioma segmentation.
• Evaluated models using multi-sequence MRI data.
• Identified challenges in pre-processing anonymization steps.

Innovation

• Deep-learning models for meningioma segmentation
• Multi-sequence MRI dataset with expert annotations
• Evaluation using DSC and Hausdorff Distance metrics