Electronic Theses and Dissertations

Identifier

6105

Date

2017

Document Type

Thesis

Degree Name

Master of Science

Major

Electrical and Computer Engr

Concentration

Computer Engineering

Committee Chair

Bonny Banerjee

Committee Member

Madhusudhanan Balasubramanian

Committee Member

Aaron L. Robinson

Abstract

Data is the key for developing algorithms that can infer the causes of agitation in mentally-ill individuals. Unfortunately, appropriate datasets are not readily available due to issues such as confidentiality breach and privacy invasion. This thesis addresses three problems.First, we identify a number of publicly-available datasets that are functions of actions and emotions of mentally-ill individuals. These include datasets of physiological signals, training videos and commercial movies.Second, we synthesize change in body-weight as a function of actions and emotions by extending a well-known model. The extended model is evaluated with respect to change in diet, physical activity and stress, using data reported in the literature. The evaluation results show promise.Third, we propose an efficient way to store several physiological signals with different sampling frequencies. Instead of storing every sample of each signal, a linear interpolation-based approach synchronizes the signals at the desired sampling rate as and when needed.

Comments

Data is provided by the student.

Library Comment

Dissertation or thesis originally submitted to the local University of Memphis Electronic Theses & dissertation (ETD) Repository.

Share

COinS