A recent study has explored a potential new avenue for identifying ADHD: examining the back of the eye. Published in npj Digital Medicine, the research suggests that certain characteristics in retinal fundus photographs may correlate with the disorder.
Using Retinal Images for ADHD Detection
Researchers in South Korea trained machine learning models to analyze these photographs as well as identify features associated with a professional ADHD diagnosis. The study found that higher blood vessel density, the shape and width of blood vessels, and specific changes in the optic disc were key indicators. One of the machine learning models demonstrated a high degree of accuracy (96.9%) in predicting ADHD based solely on the image analysis.
The Science Behind the Research
This research builds on the idea that brain connectivity changes linked to ADHD might also be visible in the eyes. If this proves accurate, it could lead to a faster and more reliable way to detect the disorder. The researchers, led by a team from Yonsei University College of Medicine, state in their paper that their analysis of retinal fundus photographs shows promise as a noninvasive biomarker for ADHD screening and the assessment of executive function deficits in visual attention.
Study Details and Results
The study involved 323 children and adolescents with an existing ADHD diagnosis, compared with a control group of 323 children and adolescents without the diagnosis, matched for age and sex. The AI system performed well in predicting ADHD and in identifying certain ADHD characteristics, such as impairments in visual selective attention.
Potential Advantages of this Method
While other machine learning approaches for ADHD screening exist, this method offers some advantages. It is relatively quick, simple to scale, and relies on a single data source (retinal photographs). The researchers acknowledge the need for further studies with larger and more diverse populations, including adults, as the current study focused on children and adolescents (average age 9.5 years). They also note that the AI system had difficulty differentiating autism from ADHD in additional tests.
The Importance of Early Detection
The researchers emphasize the importance of early screening and intervention for ADHD, which can improve social, familial, and academic outcomes.
