Indian Journal of Science and Technology

Abstract

Objectives: In the world of modern medicine, though there is lot of medical achievements, diseases like dementia still continue to pest human race. Early Diagnosishelps the diseased to have quality life. This study focuses on diagnosing the subject using classification technique by employing neuropsychological test. Methods: The dataset of 466 subjects are collected by conducting neuropsychological tests. The readings are recorded and compared with the cut off suggested by 10/66 Research group. The data set is preprocessed; rule based classifier is used for classification. The preprocessing techniques namely Imbalance reduction, Randomization are applied. Features are selected using Wrapper method. The data set is classified using various methods. Results: The Jrip classifier has given 100% accuracy. Conclusion: In West, some of the researchers have used neuropsychological tests with machine learning for diagnosis of Dementia. But in East, the work done is comparatively less. Most of those researchers have used MMSE and other neuropsychological batteries. The importance of this work is that the authors have used CSID battery; as it is designed by Alzheimer`s Disease International, this is tailor made battery for diagnosis. Jrip classification is applied and it fetched 100% accuracy. Future work is to compare the authors` diagnosis with 10/66 diagnosis by considering 10/66 as gold standard.

Keywords: Alzheimer`s Disease, Classification, JRip, Preprocessing, 10/66