肺图像数据库联盟图像集(LIDC-IDRI)

本数据集Lung Image Database Consortium and Image Database Resource Initiative（LIDC-IDRI）是一个公开可用的医学影像数据集，广泛用于肺部结节检测、分割与分类任务的研究。该数据集包含1018个病例，由来自7个不同医疗机构收集整理，具有良好的代表性与多样性。每个病例至少包括：一次完整的CT扫描图像（DICOM 格式）；一个XML文件，记录最多四名放射科医生对肺部结节的独立标注信息；相关元数据说明文件；数据集主要用于辅助开发和评估计算机辅助诊断（CAD）系统，特别是在肺部结节识别与良恶性判断方面。The Lung Image Database Consortium and Image Database Resource Initiative (LIDC-IDRI) dataset is a publicly available medical imaging dataset widely utilized for research on pulmonary nodule detection, segmentation and classification tasks. It comprises 1018 cases collected and curated by seven distinct medical institutions, featuring excellent representativeness and diversity. Each case includes at least: one complete computed tomography (CT) scan in Digital Imaging and Communications in Medicine (DICOM) format; an XML file recording independent annotation information of pulmonary nodules from up to four radiologists; and relevant metadata description documents. This dataset is primarily used to assist in the development and evaluation of computer-aided diagnosis (CAD) systems, particularly for pulmonary nodule identification and benign-malignant classification.

搜集汇总

背景与挑战

背景概述

LIDC-IDRI是一个公开的医学影像数据集，包含1018个病例的CT扫描图像及放射科医生的独立标注，主要用于肺部结节检测、分割与分类研究，支持计算机辅助诊断系统的开发和评估。