Computer aided diagnosis system for Alzheimers disease using principal component analysis and machine learning based approaches

TL;DR

This paper presents a computer-aided diagnosis system for Alzheimer's disease using PCA for feature extraction and machine learning classifiers like neural networks and SVMs to improve detection accuracy from MRI and PET images.

Contribution

It introduces a combined PCA and machine learning approach for AD detection, demonstrating improved accuracy over existing methods.

Findings

The system achieved high accuracy in AD classification.

PCA effectively reduced feature dimensionality.

Combined classifiers outperformed individual models.

Abstract

Alzheimers disease (AD) is a severe neurological brain disorder. It is not curable, but earlier detection can help improve symptoms in a great deal. The machine learning based approaches are popular and well motivated models for medical image processing tasks such as computer-aided diagnosis. These techniques can improve the process for accurate diagnosis of AD. In this paper, we investigate the performance of these techniques for AD detection and classification using brain MRI and PET images from the OASIS database. The proposed system takes advantage of the artificial neural network and support vector machines as classifiers, and principal component analysis as a feature extraction technique. The results indicate that the combined scheme achieves good accuracy and offers a significant advantage over the other approaches.