Brain MRI Data & Machine Learning Models Might Help In

Although Attention Deficit Hyperactivity Disorder (ADHD) is one of the most common neurological conditions that affects children and adults, it is still widely misunderstood. ADHD symptoms are commonly misdiagnosed or remain undiagnosed — particularly among girls and women.

In a new study, researchers made a breakthrough by potentially finding a far more robust mechanism through which ADHD diagnosis via brain MRI scans might become a reality in the future. A team of three researchers at the Yale School of Medicine delved into the data from MRI tests that were conducted on 7,805 children based in the United States. They managed to identify biomarkers of ADHD by using that data as vital input for machine learning models that might potentially be able to diagnose the neurological condition.

The researchers used data from the Adolescent Brain Cognitive Development (ABCD) study that investigates brain development in the U.S. on a longterm basis. The children included in the study were between the ages of 9 to 12 years. The 7805 children included by Huang Lin, a post-graduate researcher at the Yale School of Medicine and colleagues underwent MRI scans, resting-state functional MRI, and diffusion tensor imaging. Out of this group of children, 1,798 were diagnosed with ADHD.

By using statistical analysis methods of this data, the researchers were able to delve into the association of ADHD with the children’s brain volume, white matter integrity, and surface area.

Compared to children who do no have ADHD, the team found that those who were diagnosed with ADHD have abnormal connectivity in the networks of the brain that are responsible for processing memory and auditory inputs. The brain scans of children with ADHD also revealed that their brain cortex was experiencing thinning and on a microstructural level, there were significant changes in the white matter of their brains’ frontal lobe.

“The frontal lobe is the area of the brain involved in governing impulsivity and attention or lack thereof—two of the leading symptoms of ADHD,” Lin said in a press release. He further added that this data was robust enough to be fed into machine learning models that might be able to diagnose children with ADHD.

“Our study underscores that ADHD is a neurological disorder with neuro-structural and functional manifestations in the brain, not just a purely externalized behavior syndrome,” Lin added in her statement. “At times when a clinical diagnosis is in doubt, objective brain MRI scans can help to clearly identify affected children.”