"'Digital Mask' Could Protect Patients' Privacy in Medical Records"

Using three-dimensional (3D) reconstruction and deep learning algorithms, a team led by scientists from Cambridge and China was able to erase identifiable features from facial images while keeping disease-relevant features required for diagnosis. Facial images can help identify disease symptoms. Deep forehead wrinkles and wrinkles around the eyes, for example, are strongly linked to coronary heart disease. However, facial images record other biometric information about the patient, such as race, gender, age, and mood. Data breaches are becoming increasingly common as medical records become more digital. Although most patient data can be anonymized, it is more difficult to anonymize facial data while retaining essential information. Common methods, such as blurring and cropping identifiable areas, may result in the loss of important disease-relevant information. People are often hesitant to share their medical data for public medical research or electronic health records due to privacy concerns, thus slowing down the development of digital medical care. Therefore, the team created a 'digital mask,' which takes an original video of a patient's face and outputs a video based on a deep learning algorithm and 3D reconstruction, while discarding as much of the patient's personal biometric information as possible. They surveyed randomly selected clinic patients to assess their attitudes toward digital masks. More than 80 percent of patients believed the digital mask would alleviate their privacy concerns, and they showed a greater willingness to share personal information if such a measure were implemented. This article continues to discuss the digital mask developed to allow facial images to be stored in medical records while preventing potentially sensitive personal biometric information from being extracted and shared.

University of Cambridge reports "'Digital Mask' Could Protect Patients' Privacy in Medical Records"