digital mask protects patient privacy
photo: vcg
by staff reporters
chinese researchers have developed a new technology called the digital mask (dm), which is based on deep learning and real-time 3d reconstruction to remove patients identifiable features, while retaining their disease-relevant features needed for diagnosis. the technology was revealed by nature medicine recently.
they found that with the dm, examination videos of patients with manifestations of eye diseases can be precisely reconstructed from 2d videos containing original faces.
the technology is a result of joint efforts by various scientific research institutions including sun yat-sen university and tsinghua university. the researchers did an experimental study which focuses on four pathological manifestations of eyes, that is, thyroid-associated orbitopathy, strabismus, ptosis and nystagmus.
deep learning achieved feature extraction from different facial parts, and 3d reconstruction automatically digitalized the shapes and motions of 3d faces eyelids and eyeballs based on the extracted facial features.
the reconstruction consisted of three main stages: face reconstruction, eyelid reconstruction and eyeball reconstruction.
during the process, dm technology can retain the clinical attributes, while minimizing access to nonessential biometric information for added personal privacy in clinical practice.
a clinical diagnosis comparison showed that there is a very high diagnostic consistency between the use of original and reconstructed facial videos.
