Face recognition allows the phone to be unlocked. Could it also be able to play a much more important role in people's lives by using facial features to determine whether a person has a rare genetic disorder or not? DeepGestalt, an artificial intelligence developed by Boston-based technology company FDNA, suggests that the answer is a resounding "yes."
The algorithm is already being used by leading geneticists at over 2,000 locations in over 130 countries around the world. In a new study published in the journal Nature Medicine, researchers show how the algorithm has outperformed physicians in disease detection.
The study involved 1
"DeepGestalt is a framework for facial image analysis that can demonstrate similarities with hundreds of genetic disorders. Yaron Gurovich, Chief Technology Officer at FDNA, told Digital Trends. "It is a kind of artificial intelligence that is able to efficiently learn the relevant visual phenomena of genetic conditions and provide relevance values for [them]. It is based on the latest machine learning tools called deep learning. In practice, we use artificial neural networks to learn subtle patterns on the face and create a mathematical representation of them. DeepGestalt is like a mathematically aggregated representation of the knowledge of thousands [of] experts.
To create their system, researchers first learned to identify faces by using a generic face-set available on the Web. They then used a technique called "transfer learning" to teach the machine to stop genetic disorders. "This step is similar to teaching a human [a] a new topic," Gurovich continued. "If you know the basics – analyzing [how to] faces", it is much easier to learn special cases [such as analyzing] genetic disorders. "
As mentioned earlier, AD is already being used by clinicians in the form of a community platform called Face2Gene, which allows physicians to upload images to the platform with the permission of their patients, so that Face2Gene can help prevent possible illnesses for doctors It is estimated that 70 percent of clinical geneticists use the tool.