Attributed graph distance measure for automatic detection of attention deficit hyperactive disordered subjects

Area of Science:

• Neuroscience
• Medical Imaging
• Machine Learning

Background:

• Attention Deficit Hyperactive Disorder (ADHD) is a common childhood disorder with unknown causes.
• Functional Magnetic Resonance Imaging (fMRI) is increasingly used to study brain activity in ADHD.
• Resting-state fMRI (rs-fMRI) data offers insights into brain functional connectivity.

Purpose of the Study:

• To propose a novel framework for automatic classification of ADHD subjects using rs-fMRI data.
• To construct brain functional connectivity networks and analyze voxel activity.
• To develop a machine learning model for accurate ADHD detection.

Main Methods:

• Constructed brain functional connectivity networks from rs-fMRI data.
• Measured voxel activity based on fMRI time-series power.
• Utilized Multi-Dimensional Scaling (MDS) for dimensionality reduction.
• Employed Support Vector Machine (SVM) for classification.

Main Results:

• Achieved classification accuracy of 70.49% on training data and 73.55% on test data.
• Demonstrated higher detection rates when classifying male and female subjects separately.
• The proposed framework shows significant promise for ADHD classification.

Conclusions:

• The novel framework effectively classifies ADHD subjects using rs-fMRI data.
• The method's performance is promising, with potential for clinical application.
• Gender-specific analysis may improve ADHD detection accuracy.