by A team of researchers from multiple institutions in Korea has successfully developed a deep-learning-based AI system that can accurately diagnose autism in children. The team, consisting of child and adolescent psychologists, behavioral scientists, ophthalmologists, and biomedical systems informatics specialists, tested their autism-diagnosing tool using a learning-based AI system and obtained promising results.
Autism, formally known as autism spectrum disorder, is a neurodevelopmental disorder characterized by difficulties in communication, social interaction, and repetitive movements. Previous studies have suggested that 1 in 36 individuals may have some level of autism. Early diagnosis of autism is crucial as it enables early intervention and support for individuals with the disorder.
The research team based their study on the observation that brain disorders like ADHD can lead to minor retinal abnormalities. In order to investigate whether this applies to autism as well, the team conducted an experiment.
The experiment involved training an AI system to identify patterns in the retinas of children with autism. The system was then used to scan retinal images of 958 children, half of whom had autism. Remarkably, the system accurately identified every child with autism and did not produce any false positives.
Additionally, the system provided a number score that estimated the severity of the disorder in those diagnosed. However, the system’s accuracy in determining the severity was slightly lower, ranging from 48% to 66%, compared to standardized tests like the Autism Diagnostic Observation Schedule—Second Edition.
Notably, the study only included children between the ages of 4 and 18, and thus, it remains uncertain whether the system would be equally effective in diagnosing younger children, as the retina does not fully develop until the age of 4. To address this, the research team plans to conduct further experiments.
The development of an AI-based system for autism diagnosis holds significant potential. Early and accurate diagnosis can facilitate timely interventions, enabling individuals with autism to receive the support they need. The research team’s findings provide a promising step forward in this direction. Further research and experiments will be necessary to refine and validate the system’s efficacy, particularly in younger children.
Ravina Pandya, Content Writer, has a strong foothold in the market research industry. She specializes in writing well-researched articles from different industries, including food and beverages, information and technology, healthcare, chemical and materials, etc. With an MBA in E-commerce, she has an expertise in SEO-optimized content that resonates with industry professionals.
