الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 112 |  |  |
| | | from | 112 | | |
Abstract
Current pain evaluation depends on patient’s self-reporting, which requires the patient himself to go to the hospital or healthcare center. After surgery, elderly people or children cannot properly convey their pain. There are also limitations on the frequency caregivers or doctors can check on their patients. Due to this, e-healthcare systems were developed to reduce such complications. It essentially serves the elderly people who live alone and need 24/7 continuous care, which costs large amount of money. Moreover, recently as a result of what happened to the world during the past and current year of the spread of the Covid-19 epidemic, it was necessary to have a reliable healthcare system for remote observation, especially in care homes for the elderly.
Following this direction, this thesis starts by presenting a proposed approach for automated pain detection, pain classification, and face recognition. In this approach, Gabor filter is used for feature extraction, Relief-f with Self-Adaptive Differential Evolution (SADE) are used for feature selection, and finally, either AdaBoost or KNN algorithm is used for classification. Pain intensity is measured by Prkachin and Solomon pain intensity scale which divides the pain degree into three levels: 1) no pain, 2) weak pain, and 3) strong pain. According to the conducted experimental results, the proposed approach demonstrated promising results in comparison with earlier works in the literature. The proposed approach achieved 91% accuracy for pain detection, 99.89% accuracy for face recognition, and 78%, 92%, 88% accuracy respectively for the three-levels of pain classification. The thesis also introduces a smart healthcare framework based on the aforementioned proposed approach, called Remote in-Home Health Monitoring (RHHM). This framework provides various functionalities in order to facilitate the control and observation of patients’ status when they are far away from hospitals and living in their own home or living in elderly health-care homes. The framework exploits the benefits of fog layer with high-level services such as local storage, local real-time data processing, and embedded data mining for taking responsibility for handling some burdens of the sensor network and the cloud and to become a decision maker. In addition, it incorporates a camera with body sensors in diagnosis for more reliability and efficiency with privacy preserving. The performance of the proposed framework was evaluated using the popular iFogSim toolkit. The results illustrated that the proposed system’s ability to reduce latency, energy consumption, network communications, and overall response time. The efforts of this work will help support the overall goal to establish a high performance, secured and reliable smart healthcare system.