mDOT works with a set of six service projects. The goal of mDOT is to create a new capability for researchers so they can discover, optimize, and deploy temporally precise mHealth interventions in real-life. Such interventions will be individualized to the moment-to-moment biopsychosocial-environmental context of each individual to directly prevent, manage or treat medical conditions. Our three TR&D’s will be conducting research and developing unique mHealth technology to realize this vision. They will be working with real-life data from collaborating projects (CPs) and working closely with CP investigators to ensure that their research is addressing real-life health research problems. To ensure that unique technological resources being developed by mDOT can be quickly adopted by a large community of health researchers, mDOT will work with a diverse group of external investigators as service project collaborators. As the goal of mDOT is to create a new mHealth technological asset in the research community, successful transition of mDOT technology to a diverse group of service projects is essential to test and enhance the research utility of mDOT. We have assembled a diverse set of service projects. As mDOT technologies become more mature, we expect rapid growth in interested service projects that will deploy mDOT technologies.
Our current selection of SP’s includes projects that work with each TR&D. These SP’s are projects that have been in discussion with mDOT investigators about their needs that are being served by the technologies being developed by mDOT. Hence, they are well-equipped to test and deploy mDOT technologies as soon as they become available.
- SP1: Adaptive Platform for Personalized Engagement (All of Us Participant Technology Systems Center)
- SP2: Socioeconomic Status, Stress, and Smoking Cessation
- SP3: Overeating in Obesity through the Lense of Passive Sensing
- SP4: SMART Weight Loss Management
- SP5: Engineering a Just-In-Time Adaptive Smoking Cessation Intervention using NCI’s QuitGuide Mobile Application
- SP6: Computational Photo-Scatterography: Unraveling Scattered Photons for Bio-Imaging
SP1: Adaptive Platform for Personalized Engagement (All of Us Participant Technology Systems Center)
Collaborating Investigator: Praduman Jain, Vibrent Health
Funding Status: U24OD023163; NIH/OD; 5/1/2017-4/30/2022
Summary: The All of Us Research Program is a key element of the Precision Medicine Initiative (PMI). Through advances in research, technology, and policies that empower patients, the PMI will enable a new era of medicine in which researchers, health care providers, and patients work together to develop individualized care. The All of Us Research Program is a historic effort to gather data from one million or more people living in the United States to accelerate research and improve health. By taking into account individual differences in lifestyle, environment, and biology, researchers will uncover paths toward delivering precision medicine.
The Participant Technology Systems Center (PTSC) develops applications and websites for volunteers to enroll in the program, provide data, and receive updates. The center also supports ongoing testing and upgrades to improve the user experience, implements innovative participant tools, and ensures the security of all participant-facing systems. SP1 provides an approach to engage, encourage, motivate, retain and sustain a nationally representative cohort of 1M+ participants, who will be monitored for 10 or more years. SP1 uses current technologies such as mobile phones, website and feature phones to achieve the objectives of the All of Us Cohort Program as well as allow for evolving consumer technologies such as wireless sensors, wearables and evolving science that balance innovation with robustness and scalability.
TR&D Service: While the All of Us program has been testing mHealth biomarkers of sleep, daily activity, and heart rate using commercially available wearables such as Fitbit activity trackers in a subset of its cohort, new sensor-derived biomarkers from TR&D3, in particular, stress, pain flares, and participant engagement scores obtained from TR&D1 are of interest to the All of Us cohort. Moreover, the rich sensor data being continually collected in order to compute the biomarkers cannot be anonymized easily and present significant privacy concerns to participants. PTSC can use privacy management tools from TR&D3 (Aim 3) for optimization of privacy-utility trade-offs before participant data is released for third party research. A successful pilot deployment can lead to the deployment of these and other novel biomarkers and privacy management techniques in a larger subset of the All of Us cohort, and advance precision medicine.
SP2: Socioeconomic Status, Stress, and Smoking Cessation
Collaborating Investigators: Cho Lam (PI) & David Wetter, University of Utah
Funding Status: R01CA190329; NIH/NCI; 7/1/15 – 6/30/21
Summary: SP2 is examining the influence of demographics and social history, biobehavioral and psychosocial predispositions, contextual and environmental factors, and acute individual and contextual precipitants on stress, smoking lapse, and abstinence among 300 smokers from low socioeconomic status (SES) who are attempting to quit (evenly split between African Americans, Latinos, and Whites). Participants are being monitored using mobile physiological sensors to passively and objectively measure stress and smoking lapse. GPS tracking paired with self-report and sensor data can be associated with spatially- and temporally-relevant characteristics of the built environment (e.g., tobacco outlets) as well as area-level characteristics (e.g., poverty, racial composition) using geographic information system (GIS) data. Principal outcomes of interest are stress and lapse ascertained in real time through sensors, and early and long-term abstinence from smoking. Thus, key pathways can be generated that link distal predictors including demographic and social history factors, biobehavioral and psychological predispositions, neighborhood and built environment characteristics, acute momentary precipitants (e.g., discrimination, craving, self-efficacy), stress, and lapse. Similar to CP5, SP2 is recruiting daily smokers who are interested in quitting but recruiting from three major ethnicities. Data collection is ongoing; over 200 participants have already completed.
TR&D Service: Since SP2 uses a similar setup as CP5, it can be used to evaluate the ease of deployment of TRD1 technologies in projects involving daily smokers without the need for close collaboration. SP3 is recruiting and collecting data in Houston (in Texas); CP5 is recruiting participants in Salt Lake City (in Utah). SP2 is ideally suited to disseminate the novel analytics methods of TR&D1 that can be used to analyze its dense multimodal sensor and self-report data to understand the dynamics of SES, daily behaviors, and physical environment and their impact on lapse likelihood. This will be an ideal service project to test and deploy the models of lapse likelihood and multiscale predictive phenotypes in the context of smoking cessation developed in collaboration with CP5 and test its applicability in low SES population across three ethnicities.
SP3: Overeating in Obesity through the Lense of Passive Sensing
Collaborating Investigator: Nabil Alshurafa, Northwestern University
Funding Status: 1K25DK113242-01A1; NIH/NIDDK; 1/1/2018-11/30/2022
Summary: Obesity, caused primarily by overeating relative to need, is a preventable chronic disease that exacts staggering healthcare costs. SP4 aims to identify eating patterns via wearable sensor data that characterize episodes of excess calorie intake. This approach may facilitate detection of known (e.g., emotional eating, impulsive eating in response to cues, hedonic eating) and novel eating phenotypes that have different treatment implications. Being able to detect overeating objectively and passively, learning to predict it, and then intervene in a manner that adapts to an individual’s problematic overeating profile paves the way toward personalized behavioral medicine interventions. In particular, the promise of real time sensing combined with machine-learning based detection models is the ability to measure eating objectively (minimizing self-report bias), precisely (with high sensitivity and specificity), passively (without burden or disruption), and dynamically (detecting rapidly changing states).
However, one of the primary obstacles to carrying out this work is the need to obtain high-quality labeled data regarding when eating episodes occur so that supervised machine learning methods can be used to detect the occurrence of eating. To this end, SP3’s research focuses on simultaneous collection of multi-modal sensor data along with video data from a body-worn camera. The advantage of this approach is that the video data can be synchronized with the sensor data and labeled post-hoc by trained study personnel, resulting is accurate labels for learning eating detection models. The disadvantage of this approach is that it requires extensive effort on the part of the study personnel to review and annotate many hours of data from each study subject.
TR&D Service: To help reduce the effort required to accurately label data, Dr. Alshurafa will integrate uncertainty modeling methods produced by TR&D1 Aim1 into a graphical interface for video-driven labeling that his team is developing. Models provided by TR&D1 will be iteratively learned as data are labeled. To help speed the labeling process, the learned model at any stage can be used to determine unlabeled regions of video with the highest predicted label uncertainty and annotators can focus their labeling efforts on these regions. This is a form of generalized active learning where uncertainty models are used to issue suggestions for regions of interest that annotators can then refine as needed to ensure that SP3’s requirements for precise segmentation and labeling are met.
SP4: SMART Weight Loss Management
Collaborating Investigators: Inbal Nahum-Shani (collaborator), University of Michigan, and Bonnie Spring (CO-PI), Northwestern University.
Funding Status: R01 DK108678 NIH/NIDDK; 2016-2021
Summary: Obesity’s high prevalence and costs make it a public health crisis, but standard of care treatment impedes uptake and depletes resources by taking a one-size-fits-all approach. Guidelines recommend provision of expensive, burdensome treatment components (e.g., counseling, meal replacement) continuously to all consumers regardless of weight loss response. Stepped care that tries less costly evidence-based treatments first, reserving more resource-intensive treatments for suboptimal responders is a logical, equitable population health management strategy. However, stepped care approaches to obesity treatment have not yet incorporated inexpensive, widely available mHealth tools. The potential pitfall of beginning with mHealth treatment is that long-term outcome may be poor if nonresponse to initially insufficient treatment allows demoralization to set in. To reduce that risk, SP4 identifies nonresponders earlier than previously has been possible by applying a predictive model derived from its prior mHealth obesity research and quickly reallocates nonresponders to augmented treatment. The overall objective of this study is to determine the best way to sequence the delivery of mHealth tools and traditional treatment components in a stepped program of obesity treatments. By sequentially delivering treatment components based on participant response and by the use of an innovative experimental approach, the Sequential Multiple Assignment Randomized Trial, this study permits achievement of the target outcome, weight loss, with least resource consumption and participant burden.
TR&D Service: SP4 is interested in the personalization algorithms proposed by TR&D2, in particular the use of these stochastic algorithms to determine the randomization probabilities in the MRT. If the TR&D algorithms under Aims 1 and 2 are demonstrated to be robust, SP4 would be interested in including extra participants so as to conduct a feasibility study for use in informing SP4’s future research.
SP5: Engineering a Just-In-Time Adaptive Smoking Cessation Intervention using NCI’s QuitGuide Mobile Application
Collaborating Investigator: Sherine El-Toukhy, NHLBI
Note: Dr. El-Toukhy recently became a Stadtman Investigator at the National Institute on Minority Health and Health Disparities. As this position provides the equivalent of an R01 funding from the intramural division of NIH annually, she gave up the above K99. She is using her intramural funding to continue this project.
Funding Status: K99MD011755; NIH/NIMHD; 8/1/2017 to 7/31/2022
Summary: Smoking prevalence remains higher than the national average among young adults, 18-29 years old. Existing mobile-based interventions, that can offer 24/7 access to personalized content, accurate and continuous data recording, and timely feedback, are thus far are unidirectional and generic or minimally tailored to baseline individual-level characteristics of smokers. SP5 is delivering a just-in-time adaptive intervention (JITAI) via the National Cancer Institute (NCI)’s QuitGuide smoking cessation mobile application. The intervention provides personalized support to each user in real time. Type of support is stage-matched (e.g., cessation, post-cessation), and timing and frequency of support are based on baseline characteristics (e.g., smoking intensity) and continuous assessment of dynamic individual (e.g., craving) and contextual (e.g., proximity to a tobacco retailer) factors that affect cessation over the course of the intervention. The intervention is harnessing advanced features of mobile phones by integrating carbon monoxide tracking, smoking cues training, and performance-based incentives. SP5 proposes a micro-randomized trial to examine hypothesized main, interaction, mediation, and moderation effects. SP5 will shape a new version of QuitGuide that will be scaled up and released through NCI’s SmokeFree.Gov platform with the potential for high clinical impact.
TR&D Service: SP5 is interested in the personalization algorithms proposed by TR&D2, in particular, the use of these stochastic algorithms to determine the randomization probabilities in the MRT. If the TR&D algorithms under Aims 1 and 2 are demonstrated to be robust, SP5 would be interested in including extra participants so to conduct a feasibility study for use in informing SP5’s future research.
SP6: Computational Photo-Scatterography: Unraveling Scattered Photons for Bio-Imaging
Collaborating Investigator: Ashutosh Sabharwal (PI), Rice University
Funding Status: IIS-1730574; NSF/IIS; 03/01/2018 - 02/28/2023
Summary: Imaging of human body at cellular-level at various depths below the skin can fundamentally impact healthcare by providing live views of cross sections of anatomy. However, current methods to achieve cellular resolution are invasive (e.g. tissue biopsy). SP6 is developing computational imaging sensors for non-invasive, deep below the skin, and at cellular-level resolutions. These sensors will enable bio-imaging deep below the skin simply by pointing a camera at any part of the body. This would put individual users at the center of their healthcare experience and make them true partners in their healthcare delivery. The health imaging devices that result from this project will act as an important pillar in the personalized medicine revolution. This research expedition also holds the potential to launch new healthcare paradigms for chronic disease management, pediatrics, low-resource healthcare, and disaster medical care.
The development of new high rate sensors, able to capture the data necessary to reconstruct the structure of the tissue deep below the skin, constitutes the most important contribution of the project. These systems and algorithms will have the potential to break the current resolution limits of noninvasive bio-imaging by nearly two orders of magnitude, enabling cellular-level imaging at depths far beyond currently possible.
TR&D Service: SP6 is developing next generation high rate sensors that needs to be integrated into mobile health infrastructure for the users to participate in their healthcare delivery effectively. TR&D3 Aim 2 will share hardware/software co-designs for high rate sensing and embedded biomarker computations with SP6 to serve as digital backends to integrate novel sensing systems developed by SP6 into mobile health workflow. In addition, machine learning algorithms developed by TR&D3 under Aim 1 for automated discovery of micromarker will be disseminated to SP6 for developing computationally-efficient version of the new biomarkers enabled by the new imaging modalities pioneered by SP6.