De-identification of Facial Features in Magnetic Resonance Images Using Deep-Leaning Technology.
Researchers
Journal
Modalities
Models
Abstract
High-resolution medical images that include facial regions can be used to recognize the subject’s face when reconstructing 3D-rendered images from 2D sequential images, which might constitute the risk of infringement of personal information when sharing data. According to the HIPAA privacy rules, full-face photographic images and any comparable image are direct identifiers and considered as protected health information. Moreover, GDPR categorizes facial images as biometric data and stipulates that special restrictions should be placed on the processing of biometric data.
Develop software that can remove DICOM headers and facial features (eyes, nose, and ears) at the 2D sliced-image level to anonymize personal information in medical images.
A total of 240 cranial MR images were used for training the deep-learning model (144, 48, and 48 for the training, validation, and test sets, respectively, from the ADNI database). To overcome the small sample size problem, we used a data augmentation technique to create 576 images per epoch. We used attention-gated U-net for the basic structure of our deep-learning model. To validate the performance of the software, we adapted an external test set comprising 100 cranial MR images from the OASIS database.
The facial features (eyes, nose, and ears) were successfully detected and anonymized in both test sets (48 from ADNI and 100 from OASIS). Each result was manually validated in both the 2D image plane and the 3D rendered images. Furthermore, the ADNI test set was verified using Microsoft azure’s face recognition AI service. By adding a user interface, we developed and distributed (via GitHub) software named ‘Deface program’ for medical images as an open-source project.
We developed deep-learning based software for the anonymization of MR images that distorts the eyes, nose, and ears to prevent facial identification of the subject in reconstructed 3D images. It could be used to share medical big data for secondary research while making both data providers and recipients compliant with the relevant privacy regulations.