Ambient Intelligence Assisted Healthcare Monitoring

Abstract

Monitoring is a process of continuously gathering data and performing real-time analysis, monitoring can improve the assessment of the current state, optimization of the business processes, identification of the critical situation, new opportunities, and assisted in decision support and planning. Traditional healthcare monitoring and services are usually offered within hospitals or medical centers, patient's vital signs measurements is conducted by traditional measurements approach. This is costly, inefficient and inconvenient for the people with the need of routine checks. Thus addressing issues pertinent to healthcare monitoring. This dissertation is focusing on the investigation of Ambient Intelligence (AmI) assisted healthcare monitoring model. The data set was constructed by simulation of patients wearable sensors from the environment of Baraha Medical City in Shambat, Khartoum North, Sudan. In connection, with this, we defined and developed proposed integrated AmI healthcare monitoring architecture framework. This research followed a research scientific design approach employing specific methods, including literature review, qualitative data analysis, and data mining (DM) techniques. Literature review was used to define theoretical background and relate the result to the knowledge in the area. In same line to this, Zachman Framework (ZF) was used to guide the development of the AmI healthcare monitoring Information Architecture (IA). Extensive investigation to develop a new novel ensemble health care decision support system for assisting an intelligent health monitoring system were carried and also focusing to reduce the dimensionality of the attributes was done. Extensive investigations of the experimental results of the performance of different Meta classifiers techniques for classifying the data from different wearable sensors used for monitoring different diseases was carried. Results have shown that the architectural representation guided by the selected framework provide a holistic view to the management of healthcare monitoring data. This IA can serve as a strategic guide to the review and development of the healthcare monitoring data collection and analysis systems. The development of AmI healthcare monitoring IA in an enterprise view in the study and design of a AmI healthcare monitoring IA is original contribution of this research, which improves and expands the conceptual framework of the research in this field. Moreover, in addition to identification of wearable sensors vital signs information requirement, the classification of patients situation through novel ensemble decision support and healthcare monitoring system using advanced data mining methods. Evaluation of the research showed that both the process and result of this research are valid and acceptable. The result of this research, mainly the AmI healthcare monitoring, will help healthcare monitoring organizations to revisit their focus of attention in drafting and implementing measures to reduce healthcare monitoring problems. Finally, result of the research can also be replicated to other developing countries with similar context.