Advancing Precision: A Comprehensive Review of MRI Segmentation Datasets from BraTS Challenges (2012-2025).

Area of Science:

• Medical Imaging
• Artificial Intelligence in Medicine
• Neuro-oncology

Background:

• The Brain Tumor Segmentation (BraTS) challenges have been pivotal in advancing automated brain tumor segmentation using MRI.
• These challenges have driven innovation in developing sophisticated algorithms for analyzing complex neuroimaging data.

Purpose of the Study:

• To provide a comprehensive review of the BraTS datasets from 2012 to 2024.
• To examine the evolution, challenges, and contributions of BraTS datasets to MRI-based brain tumor segmentation.
• To synthesize insights for researchers, clinicians, and industry stakeholders.

Main Methods:

• Systematic review of BraTS challenge datasets from 2012 to 2024.
• Analysis of dataset growth, complexity, and scope.
• Examination of pre-processing and annotation protocol refinements.
• Synthesis of findings on segmentation approach impact and limitations.

Main Results:

• BraTS datasets have progressively increased in size and complexity, incorporating improved pre-processing and annotation.
• The review elucidates the progression of dataset curation and its impact on state-of-the-art segmentation methods.
• Persisting limitations and future directions for dataset development and segmentation techniques are identified.

Conclusions:

• The BraTS datasets have significantly contributed to the field, enabling robust and clinically relevant segmentation methods.
• The evolution of BraTS datasets supports the advancement of precision medicine in neuro-oncology.
• The upcoming BraTS 2025 Challenge will further expand its focus to address broader clinical needs.