广东省人民医院乳腺磁共振多序列标准化影像及肿瘤三维标注数据集

该数据集包括乳腺磁共振（MRI）数据，包括T2, DWI, 多期DCE-MRI,ADC，以及DCE-MRI中肿瘤三维标注数据，该数据以脱敏后nii.gz形式存储。数据处理过程包括NIfTI格式转换、多期DCE-MRI重采样与配准。其主要应用于临床研究，如结合人工智能技术训练肿瘤分割模型、诊断模型、疗效评估模型、预后预测模型等。This dataset comprises breast magnetic resonance imaging (MRI) data, including T2-weighted images (T2), diffusion-weighted images (DWI), multi-phase dynamic contrast-enhanced MRI (DCE-MRI), apparent diffusion coefficient (ADC) maps, as well as 3D tumor annotation data from DCE-MRI scans. All data are stored in de-identified .nii.gz format. The data processing workflow includes NIfTI format conversion, resampling and registration for multi-phase DCE-MRI sequences. Its primary applications cover clinical research, such as training tumor segmentation models, diagnostic models, therapeutic efficacy evaluation models and prognostic prediction models by leveraging artificial intelligence technologies.

搜集汇总

背景与挑战

背景概述

该数据集由广东省人民医院提供，包含1139例乳腺磁共振多序列标准化影像（如T2、DWI、DCE-MRI和ADC）及肿瘤三维标注数据，以nii.gz格式存储，并经过脱敏、重采样和配准处理。其主要应用于人工智能驱动的临床研究，支持训练肿瘤分割、诊断、疗效评估和预后预测模型，为医疗AI企业和科研机构提供关键数据支撑。