A new AI technology can accurately identify rare genetic disorders by analyzing the shape of someone’s face, according to a recent study published in the journal Nature Medicine.
Many people with genetic disorders have recognizable facial features. For example, those with Noonan syndrome have distinctive features such as widely spaced eyes, low-set ears, and a deep groove in the area between the nose and mouth.
A team of researchers created a neural network called DeepGestalt that can identify distinctive facial characteristics associated with certain disorders.
The neural network was trained on a dataset of over 17,000 images representing more than 200 syndromes.
The AI program outperformed doctors
DeepGestalt outperformed clinicians in identifying a range of syndromes in three trials.
According to the abstract of the study,
“DeepGestalt outperformed clinicians in three initial experiments, two with the goal of distinguishing subjects with a target syndrome from other syndromes, and one of separating different genetic subtypes in Noonan syndrome.”
In the final experiment DeepGestalt was tasked with identifying the correct syndrome on 502 images of people with such conditions. DeepGestalt included the correct answer in its top 10 suggestions 91% of the time.
Yaron Gurovich, chief technology officer at FDNA, who led the research, was quoted by CNN as saying: “It demonstrates how one can successfully apply state of the art algorithms, such as deep learning, to a challenging field where the available data is small, unbalanced in terms of available patients per condition, and where the need to support a large amount of conditions is great,”
The University of Oxford’s Christoffer Nellåker believes that this technology could be very beneficial to people with genetic syndromes.
Nellåker, who was not involved in the research, told New Scientist: “The real value here is that for some of these ultra-rare diseases, the process of diagnosis can be many, many years. This kind of technology can help narrow down the search space and then be verified through checking genetic markers.”
He added, “For some diseases, it will cut down the time to diagnosis drastically. For others, it could perhaps add a means of finding other people with the disease and, in turn, help find new treatments or cures.”
Citation
“Identifying facial phenotypes of genetic disorders using deep learning,” Yaron Gurovich, Yair Hanani, Omri Bar, Guy Nadav, Nicole Fleischer, Dekel Gelbman, Lina Basel-Salmon, Peter M. Krawitz, Susanne B. Kamphausen, Martin Zenker, Lynne M. Bird and Karen W. Gripp. Nature Medicine volume 25, pages 60–64 (2019)
