A Comparative Analysis of Machine Learning Algorithms to Predict Alzheimer's Disease

Area of Science:

• Neuroscience
• Medical Imaging
• Machine Learning

Background:

• Alzheimer's disease is a major global health concern, affecting millions and causing progressive cognitive decline.
• Dementia, a primary symptom of Alzheimer's, significantly impairs thinking and reading abilities.
• Early detection of Dementia is crucial for managing Alzheimer's disease.

Purpose of the Study:

• To develop and evaluate machine learning models for predicting Dementia.
• To compare the performance of various algorithms in identifying Dementia patients.

Main Methods:

• Utilized the Open Access Series of Imaging Studies (OASIS) dataset.
• Applied and compared Support Vector Machine, Logistic Regression, Decision Tree, and Random Forest algorithms.
• Evaluated model performance with and without fine-tuning.

Main Results:

• Support Vector Machine demonstrated the highest accuracy in Dementia detection among the tested models.
• Fine-tuning improved the performance of the machine learning models.
• The developed system offers a simple approach for identifying Dementia.

Conclusions:

• Machine learning, particularly Support Vector Machine, shows significant promise for accurate Dementia detection.
• The findings suggest a potential tool for early identification and management of Alzheimer's disease.
• Further research with larger datasets could enhance the system's generalizability.