Search Here

AI is becoming an essential part of every industry. The advancements around it are proving to be instrumental in the diagnosis of some health issues like ADHD.

Researchers found substantial variations in nine brain white matter tracts across teenagers with and without attention-deficit hyperactivity disorder or ADHD after using artificial intelligence to evaluate MRI data. Study co-author Justin Huynh and the team presented the findings at the RSNA convention. The study showed how common ADHD is in the United States, impacting an estimated 5.7 million children and adolescents between the ages of 6 and 17.

The need for less subjective tests for ADHD diagnosis

In a press release, Huynh highlighted the manifestation of ADHD at an early stage and its significant effects on daily life and societal functioning. However, the researchers noted that more objective measures are necessary in the diagnosis of ADHD since it is frequently based on subjective questionnaires. 

Unlike earlier research using limited samples, this study extracted fractional anisotropy or FA data along 30 white matter tracts using DWI images. Significantly higher FA values were found in nine white matter tracts of persons diagnosed with ADHD using a deep-learning artificial intelligence model trained on FA values from 1,371 individuals and evaluated on 333 patients.

AI and ADHD

Although the idea of using AI for the diagnosis of mental health illnesses like ADHD is not new, this recent study has definitely shed light on potentially streamlining a procedure using MRI scans for doing so.

In the past, researchers have also looked into deep learning for the use of various connectome maps that can improve the accuracy of ADHD detection compared to single-scale methods.