Cognitive Impairment Progression Analysis by Comparison of Clinical and Cognitive Scales, including Visualization using Agent-Based Simulation Approach

Alzheimer"s Disease;Cognitive impairment progression;Agent-based simulation;Logistic regression;Statistics in health care;Simulation in health care;Industrial and Operations Engineering;Industrial and Systems Engineering, College of Engineering and Computer Science

Liang, XueChen, Xi ;

University of Michigan

关键词: Alzheimer";    s Disease;    Cognitive impairment progression;    Agent-based simulation;    Logistic regression;    Statistics in health care;    Simulation in health care;    Industrial and Operations Engineering;    Industrial and Systems Engineering, College of Engineering and Computer Science;   

Others  :  https://deepblue.lib.umich.edu/bitstream/handle/2027.42/136068/Cognitive%20Impairment%20Progression%20Analysis%20by%20Comparison%20of%20Clinical%20and%20Cognitive%20Scales%2c%20including%20Visualization%20using%20Agent-Based%20Simulation%20Approach.pdf?sequence=1&isAllowed=y

瑞士|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

Alzheimer’s disease has been identified as the 6th leading cause of death in United Statesin 2015. One-third seniors die with Alzheimer’s and other dementia. Only 45% of AD patientsreport being told of their diagnosis. Staging the severity of AD is needed of no delay.The;;preclinical” phase of AD is known to be a long-term progression process towards severecognitive impairment. Clinical and Cognitive scaling methods, especially Clinical DementiaRating (CDR) and Mini-mental State Examination (MMSE), are widely used in mapping thestage of cognitive impairment status. Many studies have also identified important clinical andphysiological risk factors, such as age, smoking status, blood pressure, total serum cholesterol, current impairment status, etc. To discuss a better prediction method of AD for mild cognitive impairment patients, a longitudinal study of the impact of these influential factors is essential.The objective of this research is to firstly analyze the accuracy of both methods andsecondly to propose a method that gives higher prediction accuracy. The main contribution ofthis paper is that we gave a thorough analysis on comparison of CDR and MMSE, commented on which method works better based upon personal demographic performance and brought up a pattern recognition model to predict the probability of patients reaching the unfavorable outcome, i.e. dementia, over time. We gained a method that achieves high accuracy by combining the regression model and MMSE cognitive scales. An agent-based simulation model was brought up to visualize the change of cognitive impairment status of patients over time, in various populations.