Decision Supporting Model for One-year Conversion Probability from MCI to AD using CNN and SVM.

Area of Science:

• Neurology
• Artificial Intelligence
• Medical Imaging

Background:

• Predicting Alzheimer's disease (AD) progression from Mild Cognitive Impairment (MCI) is crucial for clinical management.
• Current literature lacks robust models for predicting MCI to AD conversion within a specific timeframe.
• Deep learning techniques show promise in extracting relevant features from medical imaging for disease prediction.

Purpose of the Study:

• To propose a novel decision support model for predicting the conversion probability from MCI to AD within one year.
• To leverage deep learning and machine learning for accurate MCI to AD conversion prediction.
• To address the unmet clinical need for early Alzheimer's disease progression forecasting.

Main Methods:

• Analysis of 165 MRI scans from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database.
• Feature extraction using Convolutional Neural Networks (CNN).
• Classification of extracted features using Support Vector Machine (SVM) with various kernels (linear, polynomial, RBF).

Main Results:

• The proposed model demonstrated high classification accuracy in predicting MCI to AD conversion.
• Specific accuracies achieved were 91.0% (linear kernel), 90.0% (polynomial kernel), and 92.3% (RBF kernel).
• The model effectively utilizes MRI data for predicting disease progression.

Conclusions:

• The developed decision support model shows significant potential for clinical application in predicting Alzheimer's disease conversion.
• The integration of CNN and SVM provides an effective approach for MCI to AD prediction.
• This study contributes a valuable tool for early identification of patients at high risk of progressing to Alzheimer's disease.