Researchers use facial biometrics to identify genetic disorders

Researchers from FDNA and the University of Bonn are trying to use facial biometrics to improve their chances of diagnosing people with rare genetic diseases. FDNA is the developer of the Face2Gene app, although it recently replaced its original algorithm with a new one developed in conjunction with the University.

Although Face2Gene relies on facial biometrics, the app itself is not designed for identification. Instead, he tries to group the faces of people with the same condition together, so he can learn to recognize common traits that might lead to a diagnosis in another case.

As Wired reports, the research is based on the idea that many genetic disorders correlate with distinct facial characteristics. The problem is that while some genetic diseases (such as trisomy 21) are relatively common, others are so rare that most doctors have never encountered them in their practice, and therefore do not have the knowledge to treat them. correctly.

So was the original Face2Gene app, which could spot around 300 of the most common disorders with a high degree of accuracy. However, FDNA needed at least seven photos to train the algorithm to read a condition, and many of the rarest disorders did not have enough documentation to reach even this low threshold.

The new GestaltMatcher algorithm was introduced to Face2Gene last month, and while it’s slightly less accurate when applied to the 300 most common conditions, it works better across a much wider range of disorders. GestaltMatcher can classify around 1,000 different conditions, even when there are only two known cases. Doing so can help doctors rule out other similar conditions and allow them to reach a more accurate diagnosis.

A successful classification does not dictate a specific treatment pathway, although it may inform future research on a particular topic. Researchers had the opportunity to test the new algorithm when two unrelated children in different countries developed the same condition in 2017, leading to a report published in 2019.

“It was kind of the first time it worked,” said Peter Krawitz, scientific director of the FDNA, who is also head of the genomics institute at the University of Bonn. “We are now able to work on disorders that the system has not learned or been trained on.”

Researchers at Okayama University have previously used facial recognition to study the effects of Parkinson’s disease. The GestaltMatcher algorithm could be a valuable diagnostic tool (especially when used in conjunction with other tests), although it should be noted that the small sample sizes created issues with bias and precision. in facial recognition systems.

Source: Wired

March 9, 2022 – by Eric Weiss

Comments are closed.