Researchers at South Korea's Yonsei University College of Medicine demonstrated the use of AI to screen for and determine the severity of autism in children and teens based on images of their retinas.
The researchers said structural retinal changes are present in people with Autism Spectrum Disorder.
After training the deep learning model on images of subjects with and without autism, the AI analyzed 958 retinas of children and teens, half with autism diagnoses.
Although the AI was 100% accurate in identifying whether or not the subject had autism, it was only 48% to 66% accurate in predicting symptom severity.
From ScienceAlert
View Full Article
Abstracts Copyright © 2023 SmithBucklin, Washington, D.C., USA
No entries found