Researchers from the University of South Australia and Flinders University have found that artificial intelligence (AI) combined with electroretinogram (ERG) testing could revolutionize the diagnosis of autism spectrum disorder (ASD) in children. The findings, published in Research in Autism Spectrum Disorders, indicate that a single flash of light to the eye can provide a faster and more accurate method for diagnosing ASD.

ASD is a neurodevelopmental condition characterized by challenges in social interactions, communication, and repetitive behaviors. Currently, diagnosing ASD requires lengthy psychological assessments and reports. However, the new AI-assisted ERG test could provide clinicians with an improved and more efficient diagnostic approach.

In the study, the researchers measured retinal responses in 217 children aged 5-16 years, including 71 diagnosed with ASD and 146 children without ASD. They discovered that the retinal responses of children with ASD differed from those without the disorder. By analyzing the retinal signal’s higher frequency components, the researchers were able to identify a strong biomarker for ASD.

The study revealed that using a single bright flash of light to the right eye produced the most accurate results. By leveraging AI processing, the test’s duration was significantly reduced, with the potential to provide results within as little as 10 minutes. This streamlined approach could save parents and children the stress, time, and money associated with lengthy assessments.

Dr. Fernando Marmolejo-Ramos, a researcher from the University of South Australia, emphasizes the significance of the test for ASD diagnoses. He explains that early interventions and support are crucial for improving the quality of life for children with ASD. However, the lack of a simple test for ASD necessitates extensive psychological assessments, leading to delays in receiving a diagnosis. The AI-assisted ERG test offers an efficient alternative that can expedite the identification of ASD and ensure prompt support for affected children.

Moreover, the non-invasive nature of the test, along with its tolerability by children, makes it more accessible and comfortable for those undergoing evaluation. Dr. Paul Constable, a researcher from Flinders University, highlights the interconnectedness of the eye and the brain, making the eye a valuable avenue to gain insights into the development of the brain in individuals with ASD. He asserts that combining the ERG test with signal analysis and machine learning holds great potential for accurately classifying ASD.

The researchers acknowledge the need for further investigation and expansion of the study to include younger children and individuals with other conditions such as attention deficit hyperactivity disorder. These future steps will help determine the specificity and wider applicability of the test. Dr. Hugo Posada-Quintero, an Assistant Professor in Biomedical Engineering at the University of Connecticut, emphasizes the potential of advanced signal processing and machine learning techniques in identifying neurodevelopmental conditions like ASD.

With continued research and technological advancements, the analytic methods employed in this study could be transformed into practical tools that facilitate clinicians in efficiently screening for and diagnosing ASD and related disorders. The AI-assisted ERG test has the potential to revolutionize the diagnostic process, providing faster and more accurate assessments while alleviating burdens for parents and children alike.

*Note: