‍

Blake, Wm. The Vision of Christ. 1821.

‍

A new study finds an AI tool can detect autism spectrum disorder with 100% accuracy using retinal scans of children’s eyes.

Engineered by scientists in South Korea, the tool has been met with skepticism from autism experts who question its high accuracy.

The study involved 958 children, both with and without autism, and used a deep neural network trained on iris scans to differentiate between the two groups.

While retinal scans have shown effectiveness in screening for conditions like Alzheimer's, experts caution that the study's findings might be too preliminary and affected by unaccounted factors such as IQ levels and medication. They urge further research, emphasizing the cautious use of AI in clinical diagnoses.

Cathy Lord, distinguished professor of psychiatry at the University of California Los Angeles: “It just seems too good to be true. It seems worth trying to replicate but I'm very skeptical.”