Note to the attendees: The meeting will take place on the NIH campus (map) in Building 31 (map), 6th floor, Wing C in Room 6. Please plan to arrive at the Gateway Center (map) for security screening by 7:15 AM. It usually takes 30 minutes to clear the security line in the morning. From the Gateway Center, it can take 25 minutes to get to the meeting room (15 minutes to walk to Building 31, and an additional 10 minutes to walk to Wing C, find the elevator and make it to Room 6 on the 6th floor). Building 31 is a large building with entrance in Wing A.

Recent advances in mobile technology have opened up enormous opportunities for scientific advancement and development of new tools that may improve patients’ health and well-being. mHealth technologies offer real-time monitoring and detection of changes in health status, support the adoption and maintenance of a healthy lifestyle, provide rapid diagnosis of health conditions, and facilitate the implementation of interventions ranging from promoting patient self-care to providing remote healthcare services. To realize the potential of mHealth, however, significant innovations in computing are needed. These innovations, however, must be closely integrated and informed by close collaboration with health researchers for them to contribute to improving human health. Due to the rich and unique challenges inherent in mHealth, we also have an opportunity to advance computing methods itself in the process of advancing mHealth. A systematic discussion is needed among mHealth thought leaders to outline a grand vision for computing research in mHealth and identify the major hurdles and opportunities for advancing mHealth.
The National Science Foundation (NSF) is sponsoring and National Institutes of Health (NIH) is hosting a National Workshop on Computing Challenges in Future Mobile Health (mHealth) Systems and Applications that will bring together mHealth thought leaders in both computing and health research. This workshop has three distinct goals. Goal 1: Identify mHealth grand challenges in computing that can lead to trans-formative advances in health. mHealth holds tremendous potential to provide unprecedented visibility into the health status of individuals in their natural environment that has not been possible otherwise. The challenges range from novel sensor development, big data mobile sensor data analytics, intervention development, theoretical modeling for evaluation, and privacy & security. As an example, consider the case of just-in-time interventions to showcase the computing research challenges. Just-in-time interventions (JIT), also called ecological momentary interventions (EMI) or just-in-time adaptive interventions (JITAI) is the next evolution of personalized medicine. Current perspectives of personalized medicine focus on tailoring the intervention based on the patient's baseline condition. The intervention is tailored based on patient's genetics, socio-demographics, stage of change, or other baseline variable. JIT extends intervention tailoring beyond baseline status and adjusts or adapts the intervention over the course of the intervention as the patient's status changes. The concept of adapting treatment to the patient's current state and situation is not new; clinicians have been adapting interventions for decades in an analog manner based on clinical judgment of patient status at each visit. What is novel is the ability to adapt interventions in an automated and digital manner on a nearly continuous basis using a range of adjustment variables including current state, environmental and social context, and responses to prior intervention attempts. Although intuitively appealing as an improvement over current intervention approaches, there are numerous challenges to JIT. First, JIT requires temporally dense and precise inputs of intervention adjustment variables with minimal patient burden to provide. Passive sensor data offer promise to deliver such data but more research and development is needed to provide comprehensive and field-tested sensing of the relevant adjustment variables, and integrating and making sense of these data. Second, there is meager empirical and theoretical basis for the intervention adjustment algorithms based on these data. Extensive research is needed to understand how to adjust and adapt intervention content, timing, intensity, etc. based on the streams of near continuous data that can be used to make these adaptation. Additionally, one could predict the effects of intervention and incorporate it in the design and adaptation. In other words, the decisions for adaptation is not only based on past and current measurements, but also expected future trajectory given past measurements and a (perhaps stochastic) model of anticipated response. As this area of research advances, JIT is likely to evolve from conditional (if-then) algorithms to more sophisticated computational modeling approaches. Accomplishing JIT involves computing research at multiple levels - Sensing the health state, behavior, context, and environment; Modeling the internal and external states and their interactions; Predicting the likely outcomes of candidate interventions. This workshop will engage mHealth thought leaders in both Computing and health research to identify mHealthresearch challenges in computing. Goal 2: Identify the major barriers to engaging computing researchers in mHealth. mHealth technologies will be an integral component of future interventions to treat illness and promote health. For the health care system and patients to benefit fully from rapidly developing technological innovations, computer scientists and engineers need to engage strongly in developing the next generation of mobile health interventions. But, there are several barriers that keep the wider computing community from actively engaging in mHealth research. High Cost to Entry: Extensive resources are needed to get into mobile health research since it requires appropriate equipment and conducting studies with human participants for collection of relevant data with approval from Institutional Review Boards (IRB’s). There is a lack of large-scale open data sets (due to privacy issues and cultural issues) as is the case with other computing domains such as data mining, signal processing, etc. There is a lack of publicly available simulators that can generate synthetic data as is the case with other areas such as computer networks, sensor networks, etc. Theoretical Models: Computing is about generality, scaling to large numbers, and mathematical rigor. Generality means applicable to a variety of health conditions and potentially to other applications; but each health condition may have different requirements and establishing applicability to any one of them requires significant time and effort. Mathematical rigor means being able to prove expected performance on safety and efficacy; but no accepted mathematical models exist for health, more specifically for behavior Matchmaking: Health researchers can collect sensor data, but have issues in analyzing them due to lack of publicly available data analytic tool kits. On the other hand, computing researchers who like to work in mHealth have difficulty identifying appropriate health researchers to collaborate with. This workshop will identify other hurdles and develop action items to find solutions to remove these hurdles. Goal 3 : Establish an academic community of mHealth researchers in computing. A sense of academic community and an appropriate platform to facilitate the coming together of the community is critical to sustaining and growing the community and providing it an identity and recognition from the broader community of researchers who are outside of the mHealth research area. This is critically important for young researchers who may be close to graduating and looking for an academic job. A variety of conferences have emerged in the computing community around health. They include Wireless Health, Body Sensor Networks, BodyNets, together with several workshops associated with UbiComp, MobiHoc, MobiSys, SenSys, etc. In order to get recognition for their work, students and faculty still prefer to and regularly publish their mHealth works in a variety of existing top-tier venues, including ACM SenSys, ACM MobiSys, ACM MobiCom, ACM/IEEE IPSN, ACM UbiComp, ACM CHI, ACM KDD, etc. As a result, the computing research community working in the area of mobile health does not have a flagship forum (conference proceeding or journal) and hence are not able to gain appropriate visibility and respect within the computing community. Fragmented publishing of high-quality mHealth research in computing makes it harder for health researchers to find a venue where they can attend to learn about the best works being done and published in the area of mHealth in computing. This workshop will brainstorm on various ways in which the mHealth researchers in computing can establish an academic community to get appropriate recognition and credit for their work in the computing discipline.