Advancing Precision: A Comprehensive Review of MRI Segmentation Datasets from BraTS Challenges (2012–2024)

The Brain Tumor Segmentation (BraTS) challenges have significantly contributed to advance research in brain tumor segmentation and related medical imaging tasks. This paper provides a comprehensive review of the BraTS datasets from 2012 to 2024, highlighting their evolution, challenges, and contributions to the field of Magnetic Resonance Imaging (MRI)-based glioma segmentation. Over the years, the BraTS datasets have expanded in size, complexity, and scope, transitioning from initial datasets of 30 patient scans in 2012 to comprehensive, multi-institutional, and clinically diverse datasets, exceeding 4,500 cases in 2024. This review delves into the annotation protocols, imaging modalities, and segmentation labels employed across the datasets, along with key innovations introduced in recent challenges, such as integrating underrepresented populations, including post-treatment imaging, and focusing on algorithmic generalizability. By synthesizing insights from over a decade of BraTS challenges, this review aims to provide researchers with a detailed understanding of the datasets’ progression, their role in shaping state-of-the-art segmentation methods and their potential application to precision medicine. Results highlight the increasing utility of BraTS datasets in fostering robust and transferable segmentation models, while also identifying limitations, such as the need for improved handling of missing data and broader validation scenarios. This review concludes by emphasizing the pivotal role of BraTS challenges in enabling clinically impactful research to further enhance their applicability in diverse healthcare settings.