Diagnosis of autism spectrum disorder based on complex network features

Area of Science:

• Neuroscience
• Biomedical Engineering
• Data Science

Background:

• Autism spectrum disorder (ASD) is a neurodevelopmental condition affecting brain information processing.
• Early diagnosis of ASD is essential for managing secondary complications.
• Current clinical diagnoses may not always detect secondary issues early.

Purpose of the Study:

• To investigate the potential of EEG signal analysis for early ASD detection.
• To differentiate between individuals with ASD and neurotypical controls using specific EEG features.
• To establish a method for early identification of ASD in young children.

Main Methods:

• Utilized the visibility graph (VG) algorithm to analyze single-channel C3 electroencephalogram (EEG) signals.
• Examined topological features of complex networks derived from EEG signals.
• Employed average degree (AD) as a key metric for distinguishing between ASD and normal samples.

Main Results:

• The proposed method achieved an accuracy of 81.67% in discerning the ASD class.
• Significant differences in topological network features were observed between ASD and control groups.
• The study successfully identified patterns indicative of ASD in EEG data.

Conclusions:

• EEG signals from the C3 channel, analyzed via visibility graph topological features (like Average Degree - AD), can distinguish ASD from neurotypical controls.
• This approach offers a promising avenue for early-age diagnosis of ASD.
• The findings support the use of neurophysiological data and network analysis for identifying ASD.