Accountability in a Context-Aware Smarthome Healthcare Reasoning System

Pervasive healthcare is emerging as a possible solution for the healthcare needs of an increasingly elderly population in the developed world. It is capable of providing a technology-driven approach to alleviate the healthcare needs by allowing healthcare to move from hospital-centered care to self-care, mobile care, and at-home care. When pervasive computing is used to provide technology-driven assistive healthcare, there is a need for the system to be as intelligent and sophisticated as possible, while also being as transparent and accountable as possible for both subject and caregivers. The CARA (Context Aware Real-time Assistant) healthcare system implements a sophisticated hybrid reasoning framework that incorporates both rule-based and case-based reasoning mechanisms. This provides a novel solution that combines context awareness, general domain knowledge, and automated intelligence for pervasive healthcare. The use of a sophisticated reasoning engine in healthcare demands accountability of the system. To improve the accountability of the reasoning system, we explored the use of case provenance to provide better understanding of the reasoning outcome to the user, and increase transparency. We developed a semantic-based mechanism to check the correctness and coherence of the fuzzy rules by detecting possible conflicts, this aims to eliminate inconsistency.