Yonsei University College of Medicine will develop an artificial intelligence platform for autism screening with government support.
Yonsei University Health System announced on Thursday that a research team led by Professor Cheon Keun-ah of the Department of Child and Adolescent Psychiatry at Severance Hospital has been selected for the "2026 Public Research Outcomes Demonstration Pilot Project," organized by the Ministry of Science and ICT and the Commercialization Promotion Agency for R&D Outcomes.
The program supports the verification of laboratory-stage research outcomes in actual industrial environments and helps expand them into products for commercialization. A total of 1.32 billion won ($910,000) will be provided over 33 months, from April 2026 to December 2028.
Professor Cheon, an authority in developmental disorders including autism spectrum disorder, will serve as principal investigator. Professor Park Yu-rang of the Department of Biomedical Systems Informatics at Yonsei University College of Medicine and Healthrhy Positive, a computer programming company, will participate in the project. The ultimate goal is to develop a multimodal AI-based autism trait screening health support platform that integrates fundus images and developmental behavioral indicators, and to deploy it for actual use in medical settings and households that need it.
A fundus examination captures images of the eye using simple, camera-like equipment. The non-invasive test, which can be completed within five minutes, produces fundus images that serve as indicators of the structural and functional characteristics of the central nervous system.
The research team previously published the technology in JAMA Network Open, issued by the American Medical Association, and has filed related patents. Given the importance of early detection and treatment of autism spectrum disorder, the project focuses on further refining the developmental behavioral assessment AI that utilizes fundus images so it can be easily used at primary care clinics and at home. The team plans to raise the Technology Readiness Level (TRL), an indicator of the actual usability of a technology, to 7 or higher, reaching a stage ready for immediate use.
"In autism spectrum disorder, atypical neurodevelopment begins in early brain development after birth, so identifying it as early as possible and starting appropriate therapeutic intervention significantly affects the prognosis," Professor Cheon said. "The AI being developed is non-invasive and allows anyone to easily check within a short time, which we expect will increase the possibility of early autism screening and therapeutic intervention."
