A survey of cyber-physical system implementations of real-time personalized interventions

Advances in sensor technology and machine learning as well as the widespread use of smartphones are shifting the focus of healthcare. Emerging paradigms such as cyber-physical systems (CPSs) make possible the transition from reactive to preventive care. CPSs can be implemented to achieve effective mobile health solutions and to provide sophisticated new mechanisms to monitor an individual's state in real-time via the use of sensors and mobile devices. Despite the significant potential impact of such systems, their implementation poses a range of complex technical challenges. This article surveys the state-of-the-art in implementations of CPSs for real-time personalized interventions. A general three layer architecture which can be used to consider current implementations is first presented along with a description of its main components. We also propose a three level taxonomy in accordance with the system capabilities. Then, the principal technical challenges, human-machine interaction challenges and future directions are discussed. Fifteen of the state-of-the-art implementations are qualitatively evaluated in terms of sensor capabilities, just-in-time reaction, interruptibility, and adherence, among other characteristics. By reviewing the state-of-the-art of the systems that have been built to-date, the focus of the review is to summarize current technical challenges and future opportunities for both future CPS implementers and behavioral scientists designing CPS for personalized interventions.