Research at mHealth System Lab

At mHealth Systems Lab, we strive to make a difference in people’s lives via rigorous interdisciplinary research. Our research involves more than twenty faculty members from over ten universities. Our collaborators span a variety of disciplines (e.g., Computer Science, Electrical Engineering, Mathematics, Statistics, Psychology, Behavioral Science, Cardiology, Physiology, Public Health, etc.), making our projects highly transdisciplinary.

mHealth Systems Lab students have access to many large-scale mobile sensor datasets. There have been a wide range of deploymentsusing data generated by MD2K's software platforms over the last 10+ years. With over 2200 users, these platforms have generated over 4.7 trillion data points through over 106,000 person-days, and these numbers are growing every day. These datasets provide invaluable aid to new, current, and prospective mHealth researchers.

Our research encompasses the following research areas in mHealth:

Research Areas

Sensor-Based mHealth Markers | Sensor-Triggered Just-in-Time mHealth Interventions 
mHealth Privacy | mHealth Sensor Platforms

Sensor-Based mHealth Markers: The promise of mHealth is to provide unprecedented visibility into the physical, physiological, psychological, social, and environment state of an individual so as to discover the causes of various diseases which can be used in the development of treatments, interventions, and prevention programs. With support from NSF, we have used the real-life sensor data collected from AutoSense to develop computationally models for sensor-based continuous monitoring of stress [1,2, 3], conversations [4,5], smoking events [6,7], oral health [8], cocaine use events [9,10], and craving [11] from sensor data. These works have established the wide utility of physiological monitoring for automated detection of human behaviors in the field and enabled identification of sensor-based predictors of adverse health events, which can be used to realize sensor-based just-in-time interventions.

1. K. Hovsepian, M. al’Absi, E. Ertin, T. Kamarck, and S. Kumar. cStress: Towards a Gold Standard for Continuous Stress Assessment in the Mobile Environment, ACM UbiComp 2015.(12 pages)
2. Plarre, K., Raij, A.B., Hossain, M., Ali, A.A., Nakajima, M., al’Absi, M., Ertin, E., Kamarck, T., Kumar, S., Scott, M., Siewiorek, D., Smailagic, A., and Wittmers, L. (2011). Continuous Inference of Psychological Stress from Sensory Measurements Collected in the Natural Environment. In Proceedings of ACM IPSN. (12 pages) (Nominated for Best Paper Award)
3. Vhaduri, S., Ali, A.A., Sharmin, M., Hovsepian, K., and Kumar, S. (2014). Estimating Drivers' Stress from GPS Traces. In Proceedings of Automotive UI, pp. 1-8.
4. Rahman, M., Ali, A.A., Plarre, K., al'Absi, M., Ertin, E., and Kumar, S. (2011). mConverse: Inferring Conversation Episodes from Respiratory Measurements Collected in the Field. In Proceedings of ACM Wireless Health.(10 pages) (Nominated for Best Paper Award)
5. Bari, Rummana, Adams, Roy J., Rahman, Md. Mahbubur, Parsons, Megan Battles, Buder, Eugene H., Kumar, Santosh. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies. 2018. (27 pages)
6. N. Saleheen, A. A. Ali, S. M. Hossain, H. Sarker, S. Chatterjee, B. Marlin, E. Ertin, M. al’Absi, and S. Kumar. puffMarker: A Multi-sensor Approach for Pinpointing the Timing of First Lapse in Smoking Cessation, ACM UbiComp 2015. (12 pages)
7. Ali, A.A., Hossain, M., Hovsepian, L., Rahman, M., Plarre K., and Kumar, (2012). mPuff: Automated Detection of Cigarette Smoking Puffs from Respiration Measurements. In Proceedings of ACM IPSN. (12 pages)
8. Akther, S., Saleheen, N., Samiei, S.A., Shetty, V., Ertin, E. and Kumar, S. (2019). mORAL: An mHealth Model for Inferring Oral Hygiene Behaviors in-the-wild Using Wrist-worn Inertial Sensors. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies.
9. Hossain, M., Ali, A.A., Ertin, E., Epstein, D., Preston, K., Umbricht, A., Chen, Y., and Kumar, S. (2014). Identifying Drug (Cocaine) Intake Events from Acute Physiological Response in the Presence of Free-living Physical Activity. In Proceedings of ACM IPSN. (12 pages)
10. Kennedy, A.P., Epstein, D. H., Jobes, M. L., Agage, D., Tyburski, M., Phillips, K., Ali, A., Bari, R., Hossain, S.M., Hovsepian, K., Rahman, M., Ertin, E., Kumar, S., and Preston, K. (Accepted). Continuous In-The-Field Measurement of Heart Rate: Correlates of Drug Use, Craving, Stress, and Mood in Polydrug Users. Drug and Alcohol Dependence. (27 pages)
11. S. Chatterjee, K. Hovsepian, et. al. mCrave: Continuous Estimation of Craving During Smoking Cessation, ACM UbiComp, 2016. (12 pages)

Sensor-Triggered Just-in-Time mHealth Interventions: The sensing and inferencing of human states becomes practically useful when used in the design, development, and delivery of an intervention to improve health. As a first step towards development of sensor-triggered just-in-time interventions, we have developed methods to determine (from sensor data) when a user may be physically, cognitively, and socially available to be engaged in an intervention [1]. We find that users are least available at work and during driving, and most available when walking outside. Subsequently, we have identified from continuous measurement of stress in the natural environment that users are most stressed during driving. By using our continuous measurement of stress and detection of instantaneous driving events from GPS (e.g., braking), we have identified major factors that cause stress during driving. We have also developed visualizations for continuous time series of stress data to help develop just-in-time interventions [2]. Going forward, to identify sensor-based predictors of smoking lapse, we are conducting a field study on 75 newly abstinent smokers (as part of an R01 from the OppNet initiative at NIH) who are wearing AutoSense before and after quitting smoking. With continuous monitoring of stress, conversation, smoking, drinking, craving, and lapse events, we plan to identify vulnerable moments that precipitate lapse in a newly abstinent smoker. Automated detection of these moments on the mobile phone will then be used to determine when to deliver a just-in-time intervention on a mobile phone to prevent lapse in newly abstinent smokers [3].

1. Sarker, H., Sharmin, M., Ali, A.A., Rahman, M., Bari, R., Hossain, M., and Kumar, S. (2014). Assessing the Availability of Users to Engage in Just-in-Time Intervention in the Natural Environment. In Proceedings of ACM UbiComp. (12 pages) # Cited: 3
2. M. Sharmin, A. Raij, D. Epstein, I. Nahum-Shani, J. G. Beck, S. Vhaduri, K. Preston, and S. Kumar. Visualization of Time-Series Sensor Data to Inform the Design of Just-In-Time Adaptive Stress Interventions, ACM UbiComp 2015. (12 pages)
3. Chatterjee, S., Moreno, A., Lizotte, S. L., Akther, S., Ertin, E., Fagundes, C. P., Lam, C., Rehg, J. M., Wan, N., Wetter, D., Kumar, S. (2019) SmokingOpp: Detecting the Smoking ‘Opportunity’ Context Using Mobile Sensors. Accepted for publication in Proc. of ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies (IMWUT).

mHealth Privacy: mHealth systems have deep privacy implications. Our work on mHealth inferencing in the FieldStream project revealed that sensors once considered innocuous, such as respiration, can reveal potentially private behaviors and psychological states such as stress, conversation, smoking, or drug use events. This raises new privacy issues for sharing of mHealth data [1], since the focus of privacy research has traditionally been on protecting the identity of individuals in a group, and not on protecting the revelation of private behaviors [2,3]. As part of a new CSR project from NSF (led by UCLA, PI: Mani Srivastava), we are working towards novel approaches to share mHealth data that prevents the revelation of undesired behaviors while permitting the inferences of desired health conditions.

1. Raij, A.B., Ghosh, A., Kumar, S., Srivastava, M.B. (2011). Privacy risks emerging from the adoption of innocuous wearable sensors in the mobile environment. In Proceedings of ACM CHI. (10 pages)
2. Kotz, D, Gunter, C.A., Kumar, S., and Weiner, J. (2016). Privacy and Security in Mobile Health: A Research Agenda. In IEEE Computer, 49(6): 22-30.
3. Saleheen, N., Chakraborty, S., Ali, N., Rahman, M.M., Hossain, S.M., Bari, R., Buder, E., Srivastava, M., Kumar, S. (2016). mSieve: Differential Behavioral Privacy in Time Series of Mobile Sensor Data. In Proceedings of ACM UbiComp, pp. 706-717.

mHealth Sensor Platforms: With support from NIH’s Genes Environment & Health Initiative (GEI), my team developed the AutoSense sensor suite [1] that hosts ten sensors (ECG, respiration, skin conductance, accelerometry, temperature, alcohol, etc.), is ultra-low-power, is optimized for on-body sensing in the mobile environment, and convenient for long-term wearing. AutoSense is complemented by a robust software framework on the mobile phone that collects continuous measurements from wearable wireless sensors, processes them to make health inferences, and solicits self-reports on the phone, all in real-time. This system has been worn by 100+ human volunteers (including daily smokers, drinkers, and drug users) for 20,000+ hours in their natural environments as part of various field studies [2]. Due to its real-life successes, AutoSense has been featured in congressional reports from NIH.
         As part of an NSF Smart Health project called EasySense, we are developing a contactless physiological sensor that uses radio frequency (RF) probes to track movements of heart and lungs without any skin contact [3]. This breakthrough advancement is accomplished by using Doppler sensing, but with ultra-wideband RF probes, and use of radar techniques and compressive sensing to realize a low-power receiver that can fit on a tiny mobile device. RF sensing also makes it possible to estimate fluid build-up in the lungs that precedes heart failure in congestive heart failure (CHF) patients, providing the first-ever opportunity to non-invasively monitor worsening of lung congestion.

1. Ertin, E., Stohs, N., Kumar, S., Raij, A.B., al'Absi, M., Kwon, T., Mitra, S., Shah, S., and Jeong, J.W. (2011). AutoSense: Unobtrusively Wearable Sensor Suite for Inferencing of Onset, Causality, and Consequences of Stress in the Field. In Proceedings of ACM SenSys. (14 pages). 
2. Rahman, M., Bari, R., Ali, A.A., Sharmin, M., Raij, A., Hovsepian, K., Hossain, M., Ertin, E., Kennedy, A., Epstein, D., Preston, K., Jobes, M., Kedia, S., Ward, K., al’Absi, M., and Kumar, S. (2014). Are We There Yet? Feasibility of Continuous Stress Assessment via Wireless Physiological Sensors. In Proceedings of ACM Conference on Bioinformatics, Computational Biology, and Health Informatics (BCB), pp. 479-488
3. Gao, J., Ertin, E., Kumar, S., and al’Absi, M. (2013). Contactless Sensing of Physiological Signals Using Wideband RF Probes. Asilomar Conference on Signals, Systems, and Computers, pp. 86-90.