JMIR Publications

ABSTRACT

Background:

In the era of increasing tools for automatic image analysis in dermatology, new machine learning models require high quality image datasets. Facial image data are needed for developing models evaluating attributes such as redness (acne and rosacea models), texture (wrinkles and aging models), pigmentation (melasma, seborrheic keratoses, aging and post inflammatory hyperpigmentation) and skin lesions. De-identifying facial images is critical for protecting patient anonymity. Traditionally, journals have required facial feature concealment typically covering the eyes, but these guidelines are largely insufficient to meet ethical and legal guidelines from Health Insurance Portability and Accountability Act for patient privacy. Currently, facial feature de-identification is a challenging task given lack of expert consensus and lack of testing infrastructure for adequacy of automatic and manual facial image detection.

Objective:

To review the current literature on automatic facial de-identification algorithms and to assess their utility in dermatology use case, defined as preservation of skin attributes (redness, texture, pigmentation, lesions) and data utility.

Methods:

We conducted a systematic search using a combination of headings and keywords to encompass the concepts of facial de-identification and privacy preservation. The databases MEDLINE (via Pubmed), Embase (via Elsevier) and Web of Science (via Clarivate) were queried from inception to 5/1/2021. Studies of wrong design and outcomes were excluded during the screening and review process.

Results:

A total of 18 studies, largely focusing on general adversarial network (GANs), were included in the final review reporting various methodologies of facial de-identification algorithms for still and video images. GAN based studies were utilized due to the algorithm?s capacity to generate high quality, realistic images. Study methods were rated individually for their utility for use cases in dermatology pertaining to skin color/pigmentation and texture preservation, data utility, and human detection, by three human reviewers (Table 1). We found that most studies notable in the literature address facial feature and expression preservation while sacrificing skin color, texture, pigmentation, which are critical features in dermatology-related data utility.

Conclusions:

Overall, facial de-identification algorithms have made notable advances such as disentanglement and face swapping techniques, while producing realistic faces for protecting privacy. However, they are sparse and currently not suitable for complete preservation of skin texture, color and pigmentation quality in facial photographs. Utilizing the current advances in AI for facial de-identification summarized herein, a novel approach is needed to ensure greater patient anonymity, while increasing data access for automated image analysis in dermatology.