An Integrated Approach for Episodic Cognition AssessmentPublic Deposited
Downloadable ContentDownload PDF
- Resource Type
Dementia and Mild Cognitive Impairment (MCI) are significant health issues and a rising cost to society. They are monitored with cognitive tests during clinical appointments that are limited healthcare system capacity and patient’s ability/willingness to attend. Current cognitive tests use behavioural measures and not direct measures of cellular change enabling a patient’s ability to compensate (reminder note) to delay identification. This work presents measurement methods for cognition between clinical appointments using an integrated approach for episodic cognition assessment. Methods assess the patient during Instrumental Activities of Daily Living (IADL) within an episodic measurement framework. Electroencephalogram (EEG) / Event Related Potential (ERP) methods are presented as an emerging alternative means to detect changes in the brain. Recent consumer EEG devices make at home use a future possibility. ERP features for healthy and MCI volunteers are defined, analyzed and machine learning identified two features to distinguish the two groups with 1 False Positive and 1 False Negative error. The measurement of two IADLs is presented: Computer game play and Driving. Two games were developed and piloted with MCI volunteers showing they could indicate cognitive change. The work presents game design needs including hint and measurement subsystems. Driving is a complex task that combines executive cognitive tasks (navigation) with over-learned cognitive tasks (turn signal use). The work presents measures of driving behavior creating a driver unique signature. Machine learning techniques show that the features will allow two drivers of a shared vehicle to be distinguished from each other with an error rate as low as 1.5%. Navigational performance measures are presented for driver trip planning to indicate executive function showing a Google maps derived reference provides best performance. Turn signal use is an over learned action that is measured through detection of turn signal use from dashboard video along with GPS and map methods to determine when signals were required. The work presents big data analytics and methods to ensure the anonymity of volunteers is preserved through presentation of a k-anonymity and differential privacy methods within the data sets. The measures are combined through the episodic measurement framework for a more holistic view of the patient.
- Thesis Degree Level
- Thesis Degree Name
- Thesis Degree Discipline
- Rights Notes
- Copyright © 2017 the author(s). Theses may be used for non-commercial research, educational, or related academic purposes only. Such uses include personal study, research, scholarship, and teaching. Theses may only be shared by linking to Carleton University Institutional Repository and no part may be used without proper attribution to the author. No part may be used for commercial purposes directly or indirectly via a for-profit platform; no adaptation or derivative works are permitted without consent from the copyright owner.
- Date Created
- In Collection: