Date of Award
Fall 2018
Document Type
Thesis
Degree Name
Master of Science (MS)
Department
Computing
First Advisor
Ahamed, Sheikh Iqbal
Second Advisor
Madiraju, Praveen
Third Advisor
Guha, Shion
Abstract
With the increased accessibility to mobile technologies, research utilizing mobile technologies in medical and public health area has also increased. The efficiency and effectiveness of healthcare services are also improved by introduction of mobile technologies. Effective pain treatment requires regular and continuous pain assessment of the patients. Mobile Health or mHealth has been an active interdisciplinary research area for more than a decade to research pain assessment through different software research tools. Different mHealth support systems are developed to assess pain level of patient using different techniques. Close attention to participant’s self- reported pain along with data mining based pain level detection could help the healthcare industry and researchers to deliver effective health services in pain treatment. Pain expression recognition can be a good way for data mining based approach though pain expression recognition itself may utilize different approach based on the research study scope. Most of the pain research tools are study or disease specific. Some of the tools are pain specific (lumber pain, cancer pain etc) and some are patient group specific (neonatal, adult, woman etc). This results in recurrent but potentially avoidable costs such as time, money, and workforce to develop similar service or software research tools for each research study. Based on the pain study research characteristics, it is possible to design and implement a customizable and extensible generic pain research tool. In this thesis, we have proposed, designed, and implemented a customizable personalized pain study platform tool following a micro service architecture. It has most of the common software research modules that are needed for a pain research study. These include real-time data collection, research participant management, role based access control, research data anonymization etc. This software research tool is also used to investigate pain level detection accuracy using evidence-based continuous learning from facial expression which yielded about 71% classification accuracy. This tool is also HIPAA compliant and platform independent which makes it device independent, privacy-aware, and security-aware.