Anonymized Image Data for DWI and T2W MRI Registration Quality Assurance

The following are pre-radiotherapy T2W and DWI MRI sequences in Digital Imaging and Communications in Medicine (DICOM) format for 20 patients curated from the MD Anderson Databases (NCT03145077). <br> For each image set (T2W image and DWI image), ground truth segmentations for the left and right submandibular glands, left and right parotid glands, cervical spinal cord, brainstem, and primary gross tumor volume were manually generated by a trained physician expert (radiologist with &gt; 5 years of experience in HNC). In a subset of five cases, segmentations for all structures in both sequences were also manually generated by three additional separate observers (two physicians and one medical student). All segmentations were generated in Velocity AI (v.3.0.1; Varian Medical Systems; Palo Alto, CA, USA) in DICOM RT structure format. <br> DICOM data was anonymized using an in-house Python script that implements the RSNA CRP DICOM Anonymizer software. All files have had any DICOM header info and metadata containing PHI removed or replaced with dummy entries.

本数据集包含从MD安德森数据库（临床试验编号NCT03145077）中筛选整理得到的20例患者的放疗前T2加权成像（T2-weighted, T2W）与弥散加权成像（diffusion-weighted imaging, DWI）序列，所有数据均采用医学数字成像和通信（Digital Imaging and Communications in Medicine, DICOM）格式。
针对每一组图像集（T2加权图像与弥散加权图像），一名拥有5年以上头颈肿瘤（head and neck cancer, HNC）影像诊断经验的资深放射科医师，手动生成了左右颌下腺、左右腮腺、颈髓、脑干以及原发性大体肿瘤体积的金标准分割掩码。
在其中5例组成的子集中，上述所有结构在两组序列中的分割掩码，也由另外3名独立观察者（2名医师与1名医学生）分别手动生成。
所有分割掩码均通过Velocity AI（v.3.0.1；Varian Medical Systems；美国加利福尼亚州帕洛阿尔托市瓦里安医疗系统公司）软件，以DICOM放射治疗结构（DICOM RT Structure）格式生成。
本数据集的DICOM数据采用集成RSNA CRP DICOM匿名化器软件的自研Python脚本完成匿名化处理，所有文件中包含受保护健康信息（Protected Health Information, PHI）的DICOM头信息与元数据均已被移除或替换为虚拟条目。