tejovanth/Brain_Tumor_MRI_Dataset

该数据集包含7,023张人脑MRI图像，分为四个类别：Glioma（胶质瘤）、Meningioma（脑膜瘤）、Pituitary（垂体瘤）和No Tumor（无肿瘤）。数据集旨在用于脑肿瘤检测和分类的深度学习研究，特别是卷积神经网络（CNNs）的应用。数据来源包括Figshare、SARTAJ和Br35H，其中“无肿瘤”类别的图像来自Br35H数据集。数据集中的图像大小不一，建议在训练前进行大小调整和边缘去除。部分原始SARTAJ数据集中的胶质瘤图像标签有误，已用Figshare中的正确图像替换。推荐用途包括分类任务（预测肿瘤类型）、分割任务（识别肿瘤位置）以及多任务模型（检测、分类和定位肿瘤）。

This dataset contains 7,023 human brain MRI images organized into four categories: Glioma, Meningioma, Pituitary, and No Tumor. It is designed for research in brain tumor detection and classification using deep learning methods, particularly Convolutional Neural Networks (CNNs). The dataset is a combination of three sources: Figshare, SARTAJ, and Br35H, with the No Tumor class images sourced from Br35H. Image sizes vary, and resizing and margin removal are recommended before training. Some glioma images in the original SARTAJ dataset were incorrectly labeled and have been replaced with correct ones from Figshare. Recommended usage includes classification tasks (predicting tumor type), segmentation tasks (identifying tumor location), and multi-task models (detecting, classifying, and locating tumors).