Computer Aided Diagnosis System for multiple sclerosis disease based on phase to amplitude coupling in covert visual...

Area of Science:

• Neuroscience
• Medical Imaging
• Computational Biology

Background:

• Computer-aided diagnosis (CAD) systems are crucial for detecting neurological abnormalities and improving medical diagnosis and treatment consistency.
• Research on the efficacy of attention tasks for diagnosing multiple sclerosis (MS) using EEG signals is limited.
• Developing sensitive and specific EEG features is vital for characterizing MS patient states.

Purpose of the Study:

• To develop a novel CAD system utilizing EEG signals for diagnosing MS during covert visual attention tasks.
• To investigate the role of phase-amplitude coupling (PAC) in information encoding during visual attention in MS patients.
• To enhance the sensitivity and specificity of MS diagnosis through advanced EEG feature characterization.

Main Methods:

• Phase-amplitude coupling (PAC) of EEG signals was evaluated for MS diagnosis in healthy individuals and MS patients performing visual attention tasks.
• Machine learning algorithms were employed to classify EEG signals for disease presence.
• Feature selection using T-test and Bhattacharyya distance criteria addressed dimensionality, with leave-one-subject-out cross-validation assessing system validity.

Main Results:

• The online sequential extreme learning machine (OS-ELM) classifier combined with T-test feature selection demonstrated superior performance.
• Peak diagnostic accuracy reached 91% for the color task and 90% for the direction task.
• High sensitivity (83% color, 82% direction) and specificity (96% for both tasks) were achieved, indicating robust diagnostic capability.

Conclusions:

• The developed CAD system shows promise for the automatic diagnosis of early-stage MS.
• This approach can potentially facilitate treatment assessment in MS patients.
• The findings highlight the utility of PAC analysis in EEG for neurological disorder diagnosis.