Brain MRI Analysis for Alzheimer's Disease Diagnosis Using CNN-Based Feature Extraction and Machine Learning

Area of Science:

• Neurology
• Medical Imaging
• Artificial Intelligence

Background:

• Alzheimer's disease is a leading cause of death in older adults, with increasing mortality rates.
• Early diagnosis is crucial for improving patient survival and management.
• Current machine learning methods for Alzheimer's diagnosis using MRI require complex, expert-driven feature extraction.

Purpose of the Study:

• To propose and evaluate a deep learning model for automated Alzheimer's disease diagnosis using MRI.
• To leverage ResNet50 for automatic feature extraction, reducing reliance on manual methods.
• To compare the diagnostic performance of the proposed deep learning approach against conventional classifiers.

Main Methods:

• Utilized a pre-trained Convolutional Neural Network (CNN) model, ResNet50, for automatic feature extraction from MRI images.
• Employed the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset for model training and validation.
• Evaluated the CNN model's performance using Softmax, Support Vector Machine (SVM), and Random Forest (RF) classifiers, measuring accuracy.

Main Results:

• The ResNet50 deep learning model demonstrated superior performance in Alzheimer's disease diagnosis.
• Achieved high accuracy rates ranging from 85.7% to 99% on the MRI ADNI dataset.
• Outperformed other state-of-the-art models in diagnostic accuracy.

Conclusions:

• Deep learning, specifically ResNet50, offers an effective automated approach for Alzheimer's disease diagnosis from MRI.
• This method simplifies the diagnostic process by eliminating the need for manual feature engineering.
• The high accuracy achieved suggests significant potential for clinical application in early Alzheimer's detection.