five

Standardized representation of the TCIA LIDC-IDRI annotations using DICOM

收藏
www.cancerimagingarchive.net2020-03-26 更新2025-01-22 收录
下载链接:
https://www.cancerimagingarchive.net/analysis-result/dicom-lidc-idri-nodules/
下载链接
链接失效反馈
官方服务:
资源简介:
This dataset contains standardized DICOM representation of the annotations and characterizations collected by the LIDC/IDRI initiative, originally stored in XML and available in the TCIA <a href="/collection/lidc-idri/" target="_blank" rel="noopener">LIDC-IDRI</a> collection . Only the nodules that were deemed to be greater or equal to 3 mm in the largest planar dimensions have been annotated and characterized by the expert radiologists performing the annotations. Only those nodules are included in the present dataset.Conversion was enabled by the <em>pylidc </em>library (<a href="https://pylidc.github.io/">https://pylidc.github.io/</a>) (parsing of XML, volumetric reconstruction of the nodule annotations, clustering of the annotations belonging to the same nodule, calculation of the volume, surface area and largest diameter of the nodules) and the <em>dcmqi </em>library (<a href="https://github.com/qiicr/dcmqi">https://github.com/qiicr/dcmqi</a>) (storing of the annotations into DICOM Segmentation objects, and storing of the characterizations and measurements into DICOM Structured Reporting objects). The script used for the conversion is available at <a href="https://github.com/qiicr/lidc2dicom">https://github.com/qiicr/lidc2dicom</a>. The details on the process of the conversion and the usage of the resulting objects are available in the citation (see Citations & Data Usage Policy section).

本数据集收录了由LIDC/IDRI(肺结节图像数据库)倡议所收集的标准化DICOM格式的标注与描述信息,原始数据以XML格式存储,现可在TCIA(肿瘤图像分析数据库)的LIDC-IDRI集合中获取。仅对那些最大平面尺寸大于或等于3毫米的结节进行了专家放射科医生的标注与描述,这些结节被纳入本数据集。数据的转换得益于pylidc库(https://pylidc.github.io/)(XML解析、结节标注的体积重建、同结节标注的聚类、结节体积、表面积和最大直径的计算)以及dcmqi库(https://github.com/qiicr/dcmqi)(标注存储为DICOM分割对象,描述和测量数据存储为DICOM结构化报告对象)。用于转换的脚本可在以下链接获取:https://github.com/qiicr/lidc2dicom。有关转换过程和结果对象使用的详细信息,请参阅相关文献(参见参考文献与数据使用政策部分)。
提供机构:
The Cancer Imaging Archive
5,000+
优质数据集
54 个
任务类型
进入经典数据集
二维码
社区交流群

面向社区/商业的数据集话题

二维码
科研交流群

面向高校/科研机构的开源数据集话题

数据驱动未来

携手共赢发展

商业合作