Affiliation:
1. Department of Electronics and Communication Engineering, SRM Institute of Science and Technology College of
Engineering, Kattankulathur-603203, Tamil Nadu, India
Abstract
Background:
One of the neurodevelopmental disorders widely affecting school-aged children
in recent years is attention deficit hyperactivity disorder (ADHD). In many neurodevelopmental
disorders, grey matter may be used as a clinical indicator by looking at MRIs.
Objective:
The study aimed to segment grey matter from brain MRI using a proposed fuzzy c-means
clustering-based technique for the detection of ADHD and its subtypes (ADHD-Inattentive, ADHDHyperactive,
and ADHD-Combined). The grey matter volume, age, gender, and medication status of
the subjects were investigated to identify ADHD subtypes.
Methods:
A modified fuzzy c-means with an elbow approach has been proposed to overcome the
drawbacks of previous fuzzy c-means methods and improve segmentation performance. The volume of
segmented grey matter was included with the phenotypic information of the ADHD-200 dataset for
data analysis of typically developing (TD) and ADHD subtypes.
Results:
The proposed segmentation exhibited a dice similarity index of 95%. ADHD-Inattentive exhibited
a loss of grey matter in the prefrontal cortex, while ADHD-hyperactive exhibited a loss of grey
matter in the cerebellum when compared to TD. The analysis of ADHD subtypes based on age and
gender showed that children transitioning to adolescence are mostly affected by ADHD-inattentive and
female kids are less prone to ADHD-hyperactive. The whole grey matter volume of ADHD-inattentive
children, on average, was found to be approximately 4% less than ADHD-combined. Furthermore, the
whole grey matter volume was less in non-medication naive children.
Conclusion:
This study may support healthcare providers in giving appropriate occupational therapy
based on the identification of different ADHD subtypes.
Publisher
Bentham Science Publishers Ltd.
Subject
Radiology, Nuclear Medicine and imaging