π§ BRISC 2025
Description
BRISC 2025 β Brain Tumor MRI Dataset BRISC (BRain tumor Image Segmentation & Classification) β a curated, expert-annotated T1 MRI dataset for multi-class brain tumor classification and pixel-wise segmentation. π Published in: Scientific Data (Nature Portfolio) DOI: https://doi.org/10.1038/s41597-026-06753-y A related study with additional experiments that can serve as a baseline for comparison: π SwinβHAFNet: A Hierarchical MultiβTask Transformer for Brain Tumor Segmentation and Classification π Overview BRISC is designed to address common shortcomings in existing public brain MRI collections (e.g., class imbalance, limited tumor types, and annotation inconsistency). It provides high-quality, physician-validated pixel-level masks and a balanced multi-class classification split, suitable for benchmarking segmentation and classification algorithms as well as multi-task learning research. π¦ Dataset structure brisc2025/ ββ classification_task/ β ββ train/ β β ββ glioma/ β β β ββ brisc2025_train_00001_gl_ax_t1.jpg β β β ββ ... β β ββ meningioma/ β β ββ pituitary/ β β ββ no_tumor/ β ββ test/ β ββ glioma/ β β ββ brisc2025_test_00001_gl_ax_t1.jpg β β ββ ... β ββ meningioma/ β ββ pituitary/ β ββ no_tumor/ ββ segmentation_task/ β ββ train/ β β ββ images/ β β β ββ brisc2025_train_00001_gl_ax_t1.jpg β β β ββ ... β β ββ masks/ β β ββ brisc2025_train_00001_gl_ax_t1.png β β ββ ... β ββ test/ β ββ images/ β β ββ brisc2025_test_00001_gl_ax_t1.jpg β β ββ ... β ββ masks/ β ββ brisc2025_test_00001_gl_ax_t1.png β ββ ... ββ manifest.json ββ manifest.csv ββ manifest.json.sha256 ββ manifest.csv.sha256 ββ README.md Notes: - Classification folders contain image-level labels suitable for standard image classification pipelines. - Segmentation folders contain paired MRI images/ and corresponding binary masks/. - Image and mask filenames are identical except for file extension (images: .jpg, masks: .png). - All images are T1-weighted slices. π Dataset statistics - Total samples: 6,000 (5,000 train / 1,000 test) - Classes: 4 (balanced distribution across train/test) - Planes: Axial / Coronal / Sagittal (balanced representation) - Imaging modality: T1-weighted MRI - Annotation quality: Reviewed and corrected by medical experts π Citation This dataset is introduced in our publication: "BRISC: Annotated Dataset for Brain Tumor Segmentation and Classification" Fateh et al., 2026 If you use the BRISC dataset in your research, please cite our paper: @article{Fateh_2026, title={BRISC: Annotated Dataset for Brain Tumor Segmentation and Classification}, volume={13}, ISSN={2052-4463}, url={http://dx.doi.org/10.1038/s41597-026-06753-y}, DOI={10.1038/s41597-026-06753-y}, number={1}, journal={Scientific Data}, publisher={Springer Science and Business Media LLC}, author={Fateh, Amirreza and Rezvani, Yasin and Moayedi, Sara and Rezvani, Sadjad and Fateh, Fatemeh and Fateh, Mansoor and Abolghasemi, Vahid}, year={2026}, month=Feb }
Files
Steps to reproduce
Sources: We compiled this dataset by surveying multiple publicly available medical imaging repositories and selecting highβresolution chest and abdominal scans suitable for both classification and segmentation tasks. After an initial review of existing collections, we identified and extracted only those images that met our quality and modality criteria. All images were then crossβchecked with collaborating radiology departments to ensure clinical relevance and deβidentified in accordance with institutional privacy standards. Collection Methodology: We compiled the dataset by reviewing multiple existing medicalβimaging collections and selecting only highβquality scans suitable for both classification and segmentation. In collaboration with boardβcertified radiologists, each image was annotated at both the image and pixel levels. All cleaned and labeled images were then aggregated into a unified repository and subjected to a final audit to eliminate duplicates and ensure consistent label integrityβresulting in a precise, reliable dataset ready for research use.