Award-winning innovation: 17-year-old New Jersey student built an AI that detects autism and ADHD with a retinal scan
A school science project has turned into an award-winning innovation for 17-year-old Edward Kang, a student from New Jersey who developed an artificial intelligence tool capable of screening for autism spectrum disorder (ASD) and attention-deficit/hyperactivity disorder (ADHD) by analysing retinal images. Called RetinaMind, the experimental AI examines photographs of the back of the eye to identify subtle patterns associated with the two neurodevelopmental conditions. The project earned Kang second place and a $175,000 prize at the 2026 Regeneron Science Talent Search, one of the United States’ most prestigious competitions for young scientists. Although RetinaMind is still in the research phase, it has drawn global attention for demonstrating how AI could one day support faster and less invasive medical screening.
How RetinaMind detects autism and ADHD through a retinal scan
Edward Kang, a senior at Bergen County Academies in Hackensack, New Jersey, began working on RetinaMind in 2023 after reading research suggesting that the retina could provide clues about neurological conditions. The idea is based on the fact that the retina and the brain develop from the same embryonic tissue, meaning changes in brain development may also be reflected in the eye.To build RetinaMind, Kang trained the AI using publicly available retinal image datasets. The system analyses retinal fundus photographs, looking for microscopic patterns linked to autism, ADHD and neurotypical individuals. Unlike many artificial intelligence models that function as “black boxes”, RetinaMind also produces explainable AI heat maps, highlighting the areas of the retina that influenced its prediction. Alongside developing the AI, Kang created retinal cell models to investigate the biological reasons behind these differences, identifying several genes, including ABCA4, that may warrant further study.
How accurate is the award-winning AI?
In its initial research testing, RetinaMind achieved an 89 per cent diagnostic accuracy, a figure that attracted widespread attention after Kang presented the project at the 2026 Regeneron Science Talent Search. The competition recognised both the technical sophistication of the AI model and its effort to combine computer science with biological research, earning Kang the second-place prize of $175,000.However, the reported accuracy comes from testing on public research datasets rather than real-world hospital settings. The system has not yet undergone the large-scale clinical trials needed to confirm whether it performs consistently across different populations and healthcare environments. Researchers say these validation studies are essential before any AI-based retinal screening tool can be considered for routine medical use.
Can RetinaMind change the future of autism and ADHD screening?
Despite the excitement surrounding RetinaMind, experts emphasise that it is not a replacement for clinical diagnosis. Autism and ADHD are currently diagnosed through comprehensive evaluations that include behavioural assessments, developmental history, psychological testing and input from parents, teachers and healthcare professionals. No AI system has yet been approved to diagnose either condition using retinal scans alone.Even so, RetinaMind represents an important step in a rapidly growing area of medical research. Scientists around the world are increasingly exploring the retina as a window into brain health, with similar studies investigating conditions such as Alzheimer’s disease, Parkinson’s disease and schizophrenia. If future clinical trials validate Kang’s findings, AI-powered retinal imaging could become a valuable screening tool that helps doctors identify children who may benefit from earlier specialist assessment. For now, RetinaMind remains an experimental technology, but one that showcases how a teenager’s curiosity and scientific ambition have contributed to the future of AI-assisted healthcare.
