AI Tech Can Identify Rare Genetic Disorders By Looking at Your Face

A new study shows that some rare genetic disease can be accurately identified by a new artificial intelligence technology. It only needs a photo of the patient’s face to detect such a disease.

The tech is called DeepGestalt and, according to the study published in the journal Nature Medicine, it was better than clinicians at finding syndromes in three trials. This means that it could be valuable in personalized care.

Considering that 8% of the population has a genetic disease, and many of them are recognizable through facial features, the AI tech could help identify syndromes like Angelman – a disorder that affects the nervous system and comes with specific facial features: a wide mouth, with widely spaced teeth, strabismus, or a protruding tongue.

AI’s Deep Learning Vital in Personalized Care

The lead of the research, Yaron Gurovich, who is also the chief technology officer at FDNA (AI and precision medicine company), stated that deep learning could be used in medicine and it is successful:

“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,”

But the authors warn that facial images would be too easy to access and employers could you this system to analyze the images of potential employers and discriminate them.

For its training, researchers gave DeepGestalt 17,000 images of patients from a database where there were 200 distinct genetic syndromes.

An Immense Potential

In 91% of times, the AI proposed a list of potential syndromes and was correct for the top 10 suggestions. The AI learned facial features of the patients already diagnosed with that condition and had to identify possible disorders from the diagnoses available.

A senior lecturer in artificial medical intelligence at King’s College London, Jorge Cardoso (who wasn’t involved in the research), said that the study has its limitations. He thinks that the AI must be worked on to make sure that the algorithms are “robust in the hospital environment, clinically accurate, and applicable to different age groups and ethnic populations,” but added that “the potential of AI in healthcare is immense.”

Recommended For You

Jeff Wilkinson

About the Author: Jeff Wilkinson

Jeff Wilkinson  is a Senior Politics Reporter at Debate Report covering provincial and national politics, . Before joining  Debate Report, Jeff worked on several provincial campaigns including Jack Layton. Jeff has worked as a freelance journalist in Toronto, having been published by over 20 outlets including CBC, the Center for Media and

Leave a Reply

Your email address will not be published. Required fields are marked *