Could Artificial Intelligence Help Detect Rare Diseases Just by Looking at Faces?

January 9, 2019

CBS 

“FDNA calls their system “DeepGestalt.”It’s a form of artificial intelligence that goes through a large number of images and it learns how to classify specific genetic disorders. That’s what FDNY has now proven with a groundbreaking study in the journal Nature Medicine. DeepGestalt examined 17,000 facial images of patients diagnosed with over 200 genetic syndromes and correctly came up with a list of possible diagnoses 91 percent of the time. The facial recognition is just the beginning. FDNA says they’re working on feeding the computer videos of patient movements, speech tone and patterns as well as clinical notes and medical scans. It will integrate that with genetic information to achieve even more accurate diagnoses.”

The article investigates the potential of artificial intelligence to detect rare diseases by analyzing dysmorphic facial appearances. Researchers are utilizing AI algorithms to scan facial features for abnormalities that may indicate genetic disorders. This cutting-edge technology analyzes patterns and subtle nuances in facial structure that are often linked to rare conditions. By leveraging large datasets of facial images, the AI system enhances diagnostic accuracy and speed, enabling earlier interventions and personalized treatment plans. The article highlights how this innovative approach could revolutionize the field of rare disease diagnostics, offering hope for improved patient outcomes through the identification of dysmorphic facial appearances.