Date of Award

Spring 2015

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Mathematics, Statistics and Computer Science

First Advisor

Merrill, Stephen J.

Second Advisor

Brylow, Dennis

Third Advisor

Madiraju, Praveen

Abstract

The algorithms and mathematical methods developed in this work focus on using computational approaches for low cost solution of health care problems for better patient outcome. Furthermore, evaluation of those approaches for clinical application considering the risk and benefit in a clinical setting is studied. Those risks and benefits are discussed in terms of sensitivity, specificity and area under the receiver operating characteristics curve. With a rising cost of health care and increasing number of aging population, there is a need for innovative and low cost solutions for health care problems. In this work, algorithms, mathematical techniques for the solutions of the problems related to physiological parameter monitoring have been explored and their evaluation approaches for application in a clinical setting have been studied. The physiological parameters include affective state, pain level, heart rate, oxygen saturation, hemoglobin level and blood pressure. For the mathematical basis development for different data intensive problems, eigenvalue based methods along with others have been used in designing innovative solutions for health care problems, developing new algorithms for smart monitoring of patients; from home monitoring to combat casualty situations. Eigenvalue based methods already have wide applications in many areas such as analysis of stability in control systems, search algorithms (Google Page Rank), Eigenface methods for face recognition, principal component analysis for data compression and pattern recognition. Here, the research work in 1) multi-parameter monitoring of affective state, 2) creating a smart phone based pain detection tool from facial images, 3) early detection of hemorrhage from arterial blood pressure data, 4) noninvasive measurement of physiological signals including hemoglobin level and 5) evaluation of the results for clinical application are presented.

Share

COinS