Machine learning helps experts identify ADHD biomarkers on MRI scans

Using machine learning, experts have been able to identify several imaging features related to attention deficit hyperactivity disorder (ADHD) on MRI scans. 

Experts identified numerous white matter features, in addition to changes in white matter tracts that occur over time, which are common among people with the neurodevelopmental disorder. Given the inherently subjective nature of diagnosing ADHD, the new findings could hold significant diagnostic value, experts involved in the study suggested. [1]

“While brain imaging has been extensively used to investigate structural alterations for objective measurements in attention-deficit/hyperactivity disorder, the findings have shown considerable variation across different studies,” corresponding author Susan Shur-Fen Gau, with the department of psychiatry at National Taiwan University Hospital and College of Medicine, and colleagues explained. “Thus far, no single imaging feature has emerged as a reliable biomarker for accurately distinguishing individuals with ADHD from those without ADHD.” 

For the study, experts applied three different machine learning models to MR imaging of individuals with ADHD and a group of typically developing controls. Model 1 employed baseline white matter features of 45 white matter tracts at baseline, while Model 2 incorporated features from two time points and Model 3 included the relative rate of change per year of those same white matter tracts. 

A total of 51 patients diagnosed with ADHD and 60 typically developing controls were included.  

Model 1 yielded an area-under-the-curve (AUC) of 0.67, while Model 3, which utilized both time points and relative change per year, achieved a slightly better AUC of 0.73. At both time points, there were several white matter features observed that appeared to be common among the individuals with ADHD.  

Changes in the superior longitudinal fasciculus, frontal aslant tract, stria terminalis, inferior fronto-occipital fasciculus, thalamic and striatal tracts, and other tracts involving sensorimotor regions were also indicative of ADHD. Similarly, higher rates of change in generalized fractional anisotropy (GFA) in these areas over time correlated with improvements in visual attention, short-term memory and spatial working memory. 

The finding related to accelerated microstructural development being more common among individuals with ADHD is in line with prior neuroimaging studies on the subject, the authors noted. 

“The white matter microstructure properties and their rates of developmental changes, indicating deviations from typical developmental patterns, serve as important biomarkers for ADHD,” the group wrote. 

The study abstract is available here

Hannah murhphy headshot

In addition to her background in journalism, Hannah also has patient-facing experience in clinical settings, having spent more than 12 years working as a registered rad tech. She joined Innovate Healthcare in 2021 and has since put her unique expertise to use in her editorial role with Health Imaging.

Around the web

Richard Heller III, MD, RSNA board member and senior VP of policy at Radiology Partners, offers an overview of policies in Congress that are directly impacting imaging.
 

The two companies aim to improve patient access to high-quality MRI scans by combining their artificial intelligence capabilities.

Positron, a New York-based nuclear imaging company, will now provide Upbeat Cardiology Solutions with advanced PET/CT systems and services.