Privacy Protection in MRI Scans Using 3D Masked Autoencoders
This addresses privacy risks for patients in medical imaging by enabling high-fidelity de-identification without compromising data utility, though it is incremental as it builds on masked autoencoders.
The paper tackles the problem of protecting privacy in MRI scans by de-identifying faces through remodeling instead of removal, achieving an eight-fold increase in voxel count to 256^3 resolution and outperforming previous methods in downstream task performance and de-identification.
MRI scans provide valuable medical information, however they also contain sensitive and personally identifiable information that needs to be protected. Whereas MRI metadata is easily sanitized, MRI image data is a privacy risk because it contains information to render highly-realistic 3D visualizations of a patient's head, enabling malicious actors to possibly identify the subject by cross-referencing a database. Data anonymization and de-identification is concerned with ensuring the privacy and confidentiality of individuals' personal information. Traditional MRI de-identification methods remove privacy-sensitive parts (e.g. eyes, nose etc.) from a given scan. This comes at the expense of introducing a domain shift that can throw off downstream analyses. In this work, we propose CP-MAE, a model that de-identifies the face by remodeling it (e.g. changing the face) rather than by removing parts using masked autoencoders. CP-MAE outperforms all previous approaches in terms of downstream task performance as well as de-identification. With our method we are able to synthesize high-fidelity scans of resolution up to $256^3$ -- compared to $128^3$ with previous approaches -- which constitutes an eight-fold increase in the number of voxels.