腹部肝肿瘤靶区自动勾画数据集

腹部肝肿瘤靶区自动勾画数据集主要针对智能化125I粒子植入计划设计和剂量管理软件系统中的自动组织器官勾画功能中的子功能，肝肿瘤自动分割功能的研究。该数据集来源于领域内公认可靠的公开数据集MICCAI 2017 肝脏与肿瘤分割竞赛数据集LiTS和由法国学者公开的3D-IRCADB数据集。3D-IRCADB数据集含有多个部分标签，在筛选时，选择含有肝脏标签和肿瘤标签的数据，且保证肿瘤标签在肝脏内部（即只考虑肝内肿瘤数据），筛选后除去多余标签，只保留肝脏和肿瘤标签，其中肝脏标签值设为1，肿瘤标签值设为2。该数据集包含影像数据、真实标签、第一次预测分割结果、第二次预测分割结果的20例数据，均为CT影像数据。该数据集中共80个文件，均为.nii.gz数据，其中影像数据及真实标签为公开数据集获得，第一次预测结果及第二次预测结果通过研发的自动勾画功能分割需求，并保存下来。数据量为3.38GB。This dataset is developed for the research on the automatic liver tumor segmentation sub-function of the automatic organ segmentation module in the intelligent iodine-125 seed implantation planning and dose management software system. It is sourced from two widely recognized and reliable public datasets: the MICCAI 2017 Liver Tumor Segmentation Challenge (LiTS) dataset and the 3D-IRCADB dataset released by French scholars. The 3D-IRCADB dataset contains multiple partial labels. During screening, data with both liver and tumor labels were selected, and the tumor labels were required to be located within the liver (i.e., only intrahepatic tumor data were retained). Redundant labels were removed after screening, leaving only the liver and tumor labels, with the label value for liver set to 1 and that for tumor set to 2. This dataset includes 20 cases of CT image data, covering imaging data, ground truth labels, the first predicted segmentation results, and the second predicted segmentation results. There are a total of 80 .nii.gz format files in this dataset. The imaging data and ground truth labels were obtained from the aforementioned public datasets, while the first and second predicted segmentation results were saved in accordance with the segmentation requirements of the self-developed automatic segmentation function. The total data volume of this dataset is 3.38 GB.

搜集汇总

背景与挑战

背景概述

该数据集旨在支持肝肿瘤自动分割功能的研究，数据来源于公开的MICCAI 2017 LiTS和3D-IRCADB数据集，并经过筛选处理。它包含20例CT影像数据，共80个文件，总数据量约为3.38GB，涵盖了影像、真实标签及预测分割结果。