Three dimensional facial scans could help in the diagnosis of rare genetic diseases, according to a new study.
Researchers from the US and Canada analyzed the facial images of more than 3,000 individuals with nearly 400 different syndromes alongside images from about 700 of their relatives and 3,000 unrelated individuals. As they report this week in Genetics in Medicine, the researchers developed and trained a machine learning algorithm to detect genetic syndromes from those images. After training, the algorithm was able to discern unaffected individuals with 96 percent accuracy, and the correct diagnosis was in its top 10 ranked list for 87.2 percent of the syndromic patients.
This, the researchers say, could help diagnose patients with genetic disorders more quickly and shorten the diagnostic odyssey many have to go through currently.
"Clinical genetics is labor-intensive," senior author Ophir Klein from the University of California, San Francisco, says in a statement. "Some clinics have a two-year waiting list to get in. Using 3D imaging could dramatically enhance clinicians' ability to diagnose children more quickly and inexpensively."