How machine learning can help early detection of ADHD among kindergarten students

Artificial Intelligence could soon have a significant role in enhancing the quality of life for kids living with ADHD.

A group of researchers here at the University of Alberta discovered that machine learning can be used for early diagnosis in kids, particularly in kindergarten.

Attention Deficit Hyperactivity Disorder is one of the most common disorders in children.

The study’s first author said if left untreated, ADHD can have a huge impact on the development trajectory of kids.

“The particular reason we’re looking at young kids, there’s a huge potential. The earlier we start this kind of modelling the earlier we can capture ADHD,” explained Yang Liu, a psychiatry researcher at the University of Alberta.

The team analyzed provincial birth, health, and school records of over 23,000 children, who were in kindergarten in 2016, looking to predict which kids would be diagnosed with ADHD within the following four years.

“We believe that we could utilize data as well as machine learning to help build predictive models to identify mental health issues early, therefore enabling personalized treatments later.”

According to Liu and his team at the U of A computational psychiatry lab, some of the top predictive factors for a higher risk of ADHD are inattentive classroom behaviour, a history of mental health hospital visits, maternal mental health issues, and larger household size.

“The next steps would be improving the time period of monitoring for ADHD as well as building new models that will look at potentially vulnerable populations,” said Liu.

While his research is still in the early stages Liu says, he hopes it can help provide timely and accurate diagnostic insights.