Researchers have created a "digital mask" they say could help keep patients' images private in EHRs and during telehealth visits, according to a study published in Nature Medicine.
The digital mask, developed by researchers from the University of Cambridge (England) and China, puts out a 3D reconstruction of the patient's face using deep-learning algorithms to erase identifiable features such as race, sex and age. The Sept. 15 study found the technology worked better than the current deidentifying method — cropping images — by a 64 percent margin.
Facial images are used to diagnose conditions such as coronary heart disease and visual problems. Researchers found that ophthalmologists in the study were able to diagnose eye problems using the masked images at a comparable rate to the original videos.
"This could make telemedicine — phone and video consultations — much more feasible, making healthcare delivery more efficient," said Patrick Yu-Wai-Man, PhD, a neuroscience professor at the University of Cambridge, in a Sept. 15 university news release. "If telemedicine is to be widely adopted, then we need to overcome the barriers and concerns related to privacy protection."