In 2012, Moti Shniberg sold his facial recognition startup to Facebook and started looking for a new challenge. “We wanted to use our expertise and do something good,” he says.
Then he met the head of a medical genetics center, who explained how difficult it is to diagnose rare genetic disorders in children. Specialists sometimes use the shape and appearance of a child’s face as a clue, because some conditions, such as Down syndrome, give a child’s face a distinctive appearance. However, for many other diseases, the symptoms are more subtle and the cases are very rare.
That made Shniberg’s mind beat faster. “We knew right away that we could really help,” he says.
In 2014, Shniberg’s new startup, FDNA, launched an app called Face2Gene. It is built around a machine learning algorithm like the one he previously used to recognize individuals. Only FDNA’s algorithm analyzes a face to suggest genetic conditions a person might have.
Face2Gene is now used by thousands of geneticists worldwide. The core algorithm can identify approximately 300 conditions on a patient’s face with high accuracy. That’s a boon for geneticists and families seeking a diagnosis, but the facial algorithm still can’t see most genetic conditions. For the rarest, FDNA doesn’t have the seven or more photos of different patients needed to train the algorithm to detect the condition.
Last month, scientists at FDNA and several international institutions published results of a new algorithm called GestaltMatcher, which they say can distinguish about 1,000 disorders — an approximate three-fold increase over FDNA’s original algorithm. It is now available in the Face2Gene app.
A tricky case from 2017 helped prove the new approach. Two unrelated families in Norway and Germany each sought help from local doctors for a son with growth problems, tremors and an unusually triangular face. For both boys, tests for known genetic conditions turned out to be blank. Doctors in each country independently used gene sequencing to discover that each boy had a previously undescribed mutation in a gene called LEMD2†
The two teams got in touch through a site called GeneMatcher, which helps researchers find others puzzling over issues involving the same gene. The boys’ similar symptoms strongly suggested their shared rare mutation was to blame, but the researchers looked for additional evidence. They got it from a combination of conventional biology research and an experimental algorithm from researchers at the University of Bonn working with FDNA.
Laboratory studies suggested the boys’ mutation had similar effects on their cells as progeria, a fatal genetic disorder whose patients also have distinctive triangular faces. It is caused by mutations in genes with functions similar to LEMD2†
The experimental algorithm, a prototype inspired by Face2Gene, supported those findings. It has not attempted to identify a person’s specific condition in a photograph. Instead, it calculated how similar a face was to that of other patients. It reported that the Norwegian and German guys had very similar faces, although they had different ethical backgrounds. Their faces resembled children with progeria, but differed from it and other known conditions. “It helped confirm our impression that this is something new,” said Felix Marbach, a physician at the University of Heidelberg who worked on the project while at the University of Cologne. The pan-European team of researchers published their discovery of the condition in 2019. Identifying the gene didn’t open up new treatment options, Marbach says, but could lead to follow-up research that does.
The project showed that it was possible to use a facial algorithm to identify conditions where data was scarce or missing altogether. “That was pretty much the first time it worked,” said Peter Krawitz, FDNA’s Chief Scientific Officer and head of a genomics institute at the University of Bonn, Germany. “We can now work on conditions that the system has not learned or trained for.”
