A context-sensitive framework for mobile terminals for assisting Type 2 diabetes patients

Abstract

PURPOSE The aim of this research is to develop a mobile-based framework for using context information to create an intelligent environment within which applications for diabetes self-management operate. MOTIVATION Mobile applications are emerging as a preferred method for diabetes self-management. This increases the number of interruptions that the patient will get from the phone and can easily become annoying. Using context information holds a potential for enhancing usability. METHODS A prototype was constructed that uses a step-counter, calendar, microphone, camera, battery and onboard clock as sources of potentially useful context information. Controlled experiments were conducted to test the accuracy of context detection and the use of limited mobile resources. RESULTS The framework provided an efficient way of detecting contexts within an acceptable level of accuracy. The effect on mobile resources was almost insignificant; with effect on battery less than 15% and occupying less than 1% of main memory. CONCLUSION Context information can be used to enhance usability of disease management applications on mobile phones. Modelling context information using the proposed generalizable, reusable and extensible abstraction in combination with an event-driven architecture was suitable for mobile phones because it lowered the time and space complexity and also solves the problem of modelling context information from heterogeneous data sources.