Effective Use of Data Science Toward Early Prediction of Alzheimer's Disease
Authors:
- Mohamed MahyoubMMMohamed Mahyoub
- Martin RandlesMRMartin Randles
- Thar BakerThar Baker
- Po YangPYPo Yang
- 1 June 2018
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1455-1461
Abstract
This paper investigates data for 9 common Alzheimer's Disease risk factors, from three different categories; Medical History, Lifestyle, and Demography. The dataset used consists of 185 normal control, 177 early mild cognitive impairment, 161 late mild cognitive impairment and 127 Alzheimer's Disease subjects. The initial experiment had training results of 0.92 sensitivity, 0.935 specificity and 0.771 precision. However, during the test stage the final output was 0.741 sensitivity, 0.515 specificity and 0.286 precision. The results of this experiment did not give a clear classification or definite predictive value. Involving more variables and underlying data could provide a better outcome. This paper is a part of a long-term study that focuses on the classification and ranking the importance of Alzheimer's Disease risk factors using Machine Learning predictive models and classifications techniques.Keywords
This publication has 9 references indexed in Scilit:
- Advanced internet of things for personalised healthcare systems: A surveyPervasive and Mobile Computing, 2017
- Multiple density maps information fusion for effectively assessing intensity pattern of lifelogging physical activityNeurocomputing, 2017
- Ellipse fitting model for improving the effectiveness of life‐logging physical activity measures in an Internet of Things environmentIET Networks, 2016
- Improving the Validity of Lifelogging Physical Activity Measures in an Internet of Things Environment2015
- Early diagnosis of Alzheimer's disease with deep learning2014
- Deep LearningKI - Künstliche Intelligenz, 2012
- Alzheimer’s Disease: The Challenge of the Second CenturyScience Translational Medicine, 2011
- Feature Selection Based on Class-Dependent Densities for High-Dimensional Binary DataIEEE Transactions on Knowledge and Data Engineering, 2010
- Behavioural symptoms in patients with Alzheimer's disease and their association with cognitive impairmentBMC Neurology, 2010