Face Recognition Software Diagnoses Rare Disorders

Researchers have developed a facial recognition program that scans photos to determine if a child patient has a rare genetic disorder.

According to a study published in eLife, the facial recognition algorithm can identify genetic disorders such as progeria, Angelman syndrome or Down syndrom in young children by analyzing multiple photographs of a patient's face.

The Oxford University researchers who designed this analysis claim that such a diagnosis can help professionals identify the disorder with more confidence early on.

"A diagnosis of a rare genetic disorder can be a very important step," lead researcher Christoffer Nellåker said in a statement. "It can provide parents with some certainty and help with genetic counseling on risks for other children or how likely a condition is to be passed on."

A detailed diagnosis "can also improve estimates of how the disease might progress, or show which symptoms are caused by the genetic disorder and which are caused by other clinical issues that can be treated," the scientist added.

According to the study, previous work has established that 30 to 40 percent of all rare genetic disorders impact how the face forms in some way.

The researchers note that while recognition of some telltale facial features is already a big part of diagnosing some genetic disorders, some features can be remarkably subtle and easily overlooked.

The new program has been designed to search for and identify these subtle features with a very sensitive and objective eye. Using an initial archive of 2,878 images, 1,515 of which are healthy controls, the program "learned" to differentiate which features characterize a disorder with certainty, and which can be ignored.

In addition, the analysis can use any picture at all, as long as the face of the subject is fully visible. The researchers even designed the program to account for things like lighting, pose, facial expression and image quality in its assessment.

Nellåker hopes that future doctors will be able to simply take a smartphone picture of their patient and quickly come up with a diagnosis with the help of a computer program.

"This objective approach could help narrow the possible diagnoses, make comparisons easier and allow doctors to come to a conclusion with more certainty," he concluded.

The study was published in eLife on June 24.