Diagnosis of alzheimer’s and parkinson’s disease using artificial neural network

Abstract

Dementia is referred as any syndrome related to memory loss. Memory related problems severely affects the normal functioning of a human brain and the patient feels difficulty in memory, thinking, behavior and the ability to perform everyday activities. There exist various types of dementia, but the decisive types are Alzheimer’s disease (AD) and Parkinson’s disease (PD). This paper presents an Artificial Neural Network (ANN) for the diagnosis of AD and PD using Positron Emission Tomography (PET) scanned images. AD and PD mainly affects to the individuals with more than 60 years old and in this paper brain image of patients with age 50 to 98 is selected. To identify the presence of AD and PD, 1000 PET images are selected and processed. Presented is a Computer Aided Diagnosis (CAD) tool based on ANN for dataset training, testing and classification. The results for the diagnosis are generated automatically by comparing the input image with the trained samples in the PET image database. The classification categories include AD, PD and Healthy brain with a better accuracy of 93.14% compared to the other existing systems like SVM, Decision Tree (ID3) and Naïve Bayes Classifiers.