A Cloud-Based Framework for Context-Aware Intelligent Mobile User Interfaces in Healthcare Applications

Developing context-aware intelligent mobile user interfaces is one of the ambitious goals of pervasive computing. One of the most promising candidate domains for using such interfaces is the healthcare domain. Healthcare personnel normally work all day long in many different contexts and thus, they need intelligent mobile" user interfaces that can adapt to different contexts to help them perform their tasks efficiently and effectively without errors. Current research studies in the literature address aspects of this problem with no clear plan of how the different components may be integrated and specifically how adaptation rules and preferences can be developed, updated and managed. Thus, this paper presents a comprehensive framework that integrates several different functionalities to guide the development of intelligent mobile user interfaces with varying levels of complexity as required. The framework provides a definition of context that separates the system view and the complexity of context acquisition from the user view. It represents both context and adaptable mobile features in the form of a set of parameters and objects to facilitate the task of defining and modifying preferences for interface adaptation based on context. The framework considers both user-defined preferences and system generated ones based on usage analysis. Preferences are represented in different ways to help in handling inconsistencies and incompleteness. Storing user profiles and preferences on the cloud makes it possible for users to select stereotype preferences or to search for preferences of users with similar profiles. It also enables sharing preferences among applications and mobile devices and the automated update of stereotype preferences. Though this framework is general-purpose, it has been developed with an eye towards healthcare applications. Thus, its salient features have been tested empirically among healthcare personnel to assess their effectiveness and acceptability in this domain.