Ensemble Learning-Based Alzheimer's Disease Classification Using Electroencephalogram Signals and Clock Drawing Test...

Area of Science:

• Neuroscience
• Computer Science
• Medical Informatics

Background:

• Neurodegenerative diseases, particularly Alzheimer's disease (AD), pose a growing clinical challenge.
• Machine learning (ML) techniques, including ensemble learning (EL), show promise in medical diagnostics.
• The application of EL for AD diagnosis remains relatively underexplored.

Purpose of the Study:

• To investigate the efficacy of ensemble learning in improving Alzheimer's disease detection accuracy.
• To compare the performance of EL using combined electroencephalogram (EEG) and clock drawing test (CDT) data against individual data sources.
• To identify key features contributing to the classification of AD from healthy controls (HC).

Main Methods:

• Utilized three distinct machine learning algorithms trained on an ensemble of features.
• Combined data from electroencephalogram (EEG) and clock drawing test (CDT) for classification.
• Performed feature analysis to determine contributions to AD vs. HC classification.

Main Results:

• Ensemble learning models demonstrated superior AD detection accuracy compared to models using only EEG or CDT data independently.
• The study identified specific features that are most influential in distinguishing between AD patients and healthy controls.
• Achieved improved classification performance by integrating multimodal data within an ensemble framework.

Conclusions:

• Ensemble learning presents a promising approach for enhancing the accuracy of Alzheimer's disease detection.
• The integration of EEG and CDT data via EL offers a novel strategy for automated AD screening.
• This ML-based methodology could support clinical decision-making in the early identification of Alzheimer's disease.