Binary Classification of Alzheimer's disease using MRI images and Support Vector Machine

Dementia is a chronic syndrome affecting the daily functioning of a person due to complications through deterioration in thinking, behavior and memory. Alzheimer's disease (AD) is the most common cause of dementia, aggregating up to 1 million new cases in India alone each year. Magnetic Resonance Imaging (MRI) of the brain has been found to be a very useful modality for the detection of cerebral and neural related abnormalities. Feature extraction along with Machine Learning (ML) algorithms can be used for detection of AD. The main objective of this work is to process 2D MRI axial slices of the brain and hence, build a prediction model to classify AD subjects from normal control individuals. The axial slices of the brain have been taken from Kaggle database. Feature extraction is done using Gray Level Co-occurrence Matrix (GLCM) and Haralick features in spatial domain. The features are given to Support Vector Machine (SVM) for classification. A comparative study using other supervised ML algorithms, namely k-Nearest Neighbors (k-NN), Random Forest and Linear regression classifiers were done. The accuracy achieved using 13 Haralick features for two level classification is 84%.