AI tool could detect ADHD risk in children years before diagnosis, study finds

Researchers at Duke University Medical Center say an artificial intelligence tool using routine electronic health records can predict ADHD risk in children years before diagnosis, helping enable earlier support and intervention

According to the study, the AI system identifies children at risk of attention-deficit/hyperactivity disorder by scanning and analysing routine electronic health records for characteristic behavioural and medical patterns.

By identifying early ADHD patterns, the system potentially allows doctors to intervene sooner and provide support at a critical stage of development.

The research is published in Nature Mental Health.

Spotting ADHD early: The power of AI prediction

ADHD affects millions of children worldwide, yet many go without a diagnosis, missing out on early intervention and support. ADHD impacts a person’s ability to pay attention, maintain high energy levels, and control their impulses.

The researchers found that artificial intelligence tools can analyse routine electronic medical records to estimate a child’s risk of developing ADHD years before a typical diagnosis. By reviewing patterns in these records, the AI identifies children who may need earlier evaluation.

“We have this incredibly rich source of information sitting in electronic health records,” said Elliot Hill, lead author of the study and data scientist in the Department of Biostatistics & Bioinformatics at Duke University School of Medicine. “The idea was to see whether patterns hidden in that data could help us predict which children might later be diagnosed with ADHD, well before that diagnosis usually happens.”

How the study worked: Analysing children’s health data

The researchers analysed health data from over 140,000 children with and without ADHD, training a specialised AI model to examine medical history from birth to early childhood. The model identified combinations of developmental, behavioural, and clinical events that appeared years before an ADHD diagnosis.

The AI tool analysed clinical and demographic data to estimate future ADHD risk in children aged 5 and older, maintaining high accuracy across patient characteristics such as sex, race, ethnicity, and insurance status.

The AI software did not diagnose ADHD, but it identifies children who may benefit from closer attention from medical professionals or an earlier referral for ADHD assessment.

“This is not an AI doctor,” said Matthew Engelhard, M.D., Ph.D., in Duke’s Department of Biostatistics & Bioinformatics, and senior author of the study. “It’s a tool to help clinicians focus their time and resources, so kids who need help don’t fall through the cracks or wait years for answers.”

Earlier identification of ADHD leads to better academic, social, and health outcomes.

“Children with ADHD can really struggle when their needs aren’t understood and adequate supports are not in place,” said study author, Naomi Davis, Ph.D., associate professor in the Department of Psychiatry and Behavioral Sciences. “Connecting families with timely, evidence-based interventions is essential for helping them achieve their goals and laying a foundation for future success.”