🧠 BRISC 2025

BRISC 2025 — Brain Tumor MRI Dataset BRISC (BRain tumor Image Segmentation & Classification) — a curated, expert-annotated T1 MRI dataset for multi-class brain tumor classification and pixel-wise segmentation. 📄 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: 👉 Swin‑HAFNet: A Hierarchical Multi‑Task Transformer for Brain Tumor Segmentation and Classification: https://shorturl.at/sQtOA 🚀 Overview BRISC is designed to address common shortcomings in existing public brain MRI collections (e.g., class imbalance, limited tumor types, and annotation inconsistency). 📦 Dataset structure brisc2025/ ├─ classification_task/ │ ├─ train/ │ │ ├─ glioma/ │ │ │ ├─ brisc2025_train_00001_gl_ax_t1.jpg │ │ │ └─ ... │ │ ├─ meningioma/ │ │ ├─ pituitary/ │ │ └─ no_tumor/ │ └─ test/ │ ├─ glioma/ │ │ ├─ brisc2025_test_00001_gl_ax_t1.jpg │ │ └─ ... │ ├─ meningioma/ │ ├─ pituitary/ │ └─ no_tumor/ ├─ segmentation_task/ │ ├─ train/ │ │ ├─ images/ │ │ │ ├─ brisc2025_train_00001_gl_ax_t1.jpg │ │ │ └─ ... │ │ └─ masks/ │ │ ├─ brisc2025_train_00001_gl_ax_t1.png │ │ └─ ... │ └─ test/ │ ├─ images/ │ │ ├─ brisc2025_test_00001_gl_ax_t1.jpg │ │ └─ ... │ └─ masks/ │ ├─ brisc2025_test_00001_gl_ax_t1.png │ └─ ... ├─ manifest.json ├─ manifest.csv ├─ manifest.json.sha256 ├─ manifest.csv.sha256 └─ 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. 📊 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 📄 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 }BRISC 2025 — 脑肿瘤磁共振成像数据集 BRISC（脑肿瘤图像分割与分类，Brain Tumor Image Segmentation & Classification）—— 一份经精心甄选、专家标注的T1加权磁共振成像（T1-weighted MRI）数据集，可用于多类别脑肿瘤分类与像素级分割任务。 📄 发表期刊：《Scientific Data》（Nature Portfolio旗下子刊） DOI：https://doi.org/10.1038/s41597-026-06753-y 可作为对比基准的相关拓展实验研究： 👉 《Swin‑HAFNet：面向脑肿瘤分割与分类的分层多任务Transformer》，链接：https://shorturl.at/sQtOA 🚀 数据集概览 BRISC旨在解决当前公开脑磁共振成像数据集存在的典型缺陷，例如类别分布失衡、肿瘤类型覆盖有限、标注一致性不足等问题。 📦 数据集结构 brisc2025/ ├─ classification_task/ 分类任务文件夹 │ ├─ train/ 训练集 │ │ ├─ glioma/ 胶质瘤 │ │ │ ├─ brisc2025_train_00001_gl_ax_t1.jpg │ │ │ └─ ... │ │ ├─ meningioma/ 脑膜瘤 │ │ ├─ pituitary/ 垂体瘤 │ │ └─ no_tumor/ 无肿瘤 │ └─ test/ 测试集 │ ├─ glioma/ 胶质瘤 │ │ ├─ brisc2025_test_00001_gl_ax_t1.jpg │ │ └─ ... │ ├─ meningioma/ 脑膜瘤 │ ├─ pituitary/ 垂体瘤 │ └─ no_tumor/ 无肿瘤 ├─ segmentation_task/ 分割任务文件夹 │ ├─ train/ 训练集 │ │ ├─ images/ 图像目录 │ │ │ ├─ brisc2025_train_00001_gl_ax_t1.jpg │ │ │ └─ ... │ │ └─ masks/ 掩码目录 │ │ ├─ brisc2025_train_00001_gl_ax_t1.png │ │ └─ ... │ └─ test/ 测试集 │ ├─ images/ 图像目录 │ │ ├─ brisc2025_test_00001_gl_ax_t1.jpg │ │ └─ ... │ └─ masks/ 掩码目录 │ ├─ brisc2025_test_00001_gl_ax_t1.png │ └─ ... ├─ manifest.json ├─ manifest.csv ├─ manifest.json.sha256 ├─ manifest.csv.sha256 └─ README.md 备注： - 分类任务文件夹包含适配标准图像分类流水线的图像级标注标签。 - 分割任务文件夹包含成对的磁共振成像图像与对应的二值掩码。 - 图像与掩码的文件名完全一致，仅文件扩展名不同（图像格式为.jpg，掩码格式为.png）。 - 所有图像均为T1加权扫描切片。 📊 数据集统计信息 - 总样本量：6000例（训练集5000例 / 测试集1000例） - 类别数：4类（训练集与测试集均保持类别分布均衡） - 扫描平面：轴位（Axial）、冠状位（Coronal）、矢状位（Sagittal），三类平面样本分布均衡 - 成像模态：T1加权磁共振成像（T1-weighted MRI） - 标注质量：经医学专家审核并修正 📄 引用说明 本数据集源自以下发表论文：《BRISC：面向脑肿瘤分割与分类的标注数据集》，Fateh等人，2026年。 若您在研究中使用BRISC数据集，请引用以下论文： @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 }