UC Irvine

The availability of mobile Internet and health devices is enabling new opportunities for health researchers to access an immense amount of digital health data ubiquitously and deliver intervention and treatment online and in real time. However, to take advantage of these opportunities, the researchers need to build systems and applications to collect these data, process them, and deliver actionable items accordingly, which requires building a mobile health (mHealth) system. mHealth solutions are tools designed to assist researchers in creating their desired mHealth systems, but these solutions often have limitations in their setup time, cost, or customization level. This dissertaiton presents ZotCare, a service platform that aims to resolve these limitations by offering ready-to-use mHealth solutions in a shared environment to reduce the cost and provide a service orchestration that can be customized to different types of study by providing tools for personalization and adaptation. ZotCare's contribution also extends to its methods in creating a data pipeline scalable to the number of users, amount of occupied storage, and the load of different types of tasks by considering the opportunities and limitations in mHealth research. This dissertation will provide an introduction and background on mHealth solutions and ZotCare's position. Then, ZotCare's service orchestration is explained to demonstrate its capabilities in creating a customizable environment. This will be followed by ZotCare's data pipeline design challenges to indicate how ZotCare can run mHealth studies at scale. In the end, to showcase ZotCare's practical contribution, several use cases of ZotCare in mHealth research will be provided, followed by a short discussion on other contributions, artifacts, and future work.