DDIT - A Tool for DICOM Brain Images De-Identification

In order to protect subject privacy in the usage of medical imaging, it is desirable that images are de-identified. Two issues must be addressed for the purpose of image de-identification. One is anonymization, and the other is image defacing with regard to images that include facial features. In this paper, a tool, DDIT, is proposed for Digital Imaging and Communications in Medicine (DICOM) brain images de-identification. This tool can flexibly anonymize private data in DICOM images, and efficiently remove the potentially identifying facial features. In addition, visual inspection showed that no brain tissue was removed during the de-identification process. The tool is published online, which can be downloaded for free to help DICOM image de-identification and promote data sharing in large scale multisite projects.