In addition to allowing an early diagnosis, the study developed by the company FDNA, leader in AI, emphasizes that the use of automated facial analysis in the detection of genetic disorders adds significant value in personalized attention, which will contribute to the improvement of treatment quality. life of the patient.
The research focuses on the use of a deep learning algorithm that uses and relates more than 17,000 facial images of patients whose diagnoses cover hundreds of different genetic syndromes.
Facial analysis technology captures, structures and analyzes complex human physiological traits and stores them in a database of more than 150,000 patients.
Given the large number of possible genetic disorders, finding the right diagnosis still remains a challenge for doctors, the authors said in a press release.
However, FDNA technical director Yaron Gurovich stressed that the results of the study "open the door to future research and applications as well as the identification of new genetic syndromes."
The data used in this study were taken from the platform promoted by the Face2Gene community, in which different physicians sent facial images of more than 200 patients.
For each image, AI suggested possible potential syndromes and, in most cases, the researchers concluded that they corresponded to the clinical diagnosis.