Young American student develops AI that diagnoses autism and ADHD through retina scanning

Behavioral assessments currently used to diagnose autism and ADHD may cease to be the primary method used by specialists following the development of the RetinaMind AI by 17-year-old student Edward Kang in the United States.

The young man was developing his science project for his high school, Bergen County Academies, in New Jersey (USA). In that process, he came across “research from the University of Hong Kong that used retinal imaging to detect autism.” Kang found it “fascinating and counterintuitive” to use retinal scanning to understand what happens in the brain. His research led him to obsessively investigate using AI, machine learning, and neural networks.

 

Ver esta publicación en Instagram

 

Una publicación compartida por Acknowledge AI by Pubity (@acknowledge.ai)

This led him to create RetinaMind, a technology with an 89 % accuracy rate that “uses deep learning models to analyze retinal images and determine whether a person has autism spectrum disorder (ASD) or attention deficit hyperactivity disorder (ADHD), or if they are neurotopic.”

The student trained the model to compare different predictions made by various AIs in order to generate an average and provide a more reliable diagnosis, using neural networks specialized in image recognition.

His innovative tool incorporates heat maps that identify which areas of the retina were key to the diagnosis. With this breakthrough, it “achieved second place and a $175,000 prize in the prestigious Regeneron Science Talent Search 2026, positioning itself as a very promising technology for the detection of autism and ADHD.”

M.Pino

Source: eleconomista

(Reference image source: Ion Fet on Unsplash)

Follow our news on Google! For current, interesting, and accurate information, click here to see all the content on Bitfinance.news. You can also find us on X/Twitter and Instagram