Teen Develops AI Tool to Diagnose Autism and ADHD via Retinal Images
Edward Kang, a 17-year-old senior from Bergen County Academies in Hackensack, New Jersey, has created an innovative AI application named RetinaMind that identifies autism spectrum disorder and ADHD through retinal imaging. At the 2026 Regeneron Science Talent Search, he secured second place and won $175,000. Inspired by research from the Chinese University of Hong Kong, Kang taught himself coding and machine learning, ultimately achieving approximately 89% accuracy with his convolutional neural network. While he identified genes such as ABCA4 associated with retinal variations in neurodiverse people, experts caution about the specificity of these traits. Kang aims to enhance his model to better assess autism severity.
Key facts
- Edward Kang is 17 years old and a senior at Bergen County Academies in Hackensack, New Jersey.
- RetinaMind diagnoses autism spectrum disorder and ADHD using retinal images.
- The tool won second place and $175,000 at the 2026 Regeneron Science Talent Search.
- Regeneron Science Talent Search is the oldest and most prestigious STEM competition for U.S. high school students.
- Kang was inspired by a study from the Chinese University of Hong Kong.
- RetinaMind uses a convolutional neural network and ensemble learning.
- The tool has an accuracy rate of about 89%.
- Kang identified a dozen candidate genes, including ABCA4, linking autism and retinal development.
Entities
Institutions
- Bergen County Academies
- Chinese University of Hong Kong
- Regeneron Science Talent Search
- Society for Science
- Kennedy Krieger Institute
- Johns Hopkins Medicine
Locations
- Hackensack
- New Jersey
- United States
- Hong Kong
- Sweden