Cigarette smoking is a leading preventable cause of death, responsible for one in five deaths annually in the United States alone. Many adult smokers want to or try to quit smoking, but, are unable to do so due to lack of personalized cessation treatments that ultimately leads to successful quitting. The primary hurdle has been the unavailability of technology that can be used to collect continuous measurements of stress contexts in field on predictors of abstinence and relapse. Without these measurements, it is difficult to understand the psychophysiological and biobehavioral stress mechanisms that may predict smoke lapse. This application proposes to use a new field deployable tool called AutoSense that was developed under NIH's GEI program to collect personal psychophysiological measures of stress and mediating social and environmental contexts in the natural environment of individuals to discover signatures of early relapse. AutoSense consists of several wireless sensors (e.g., ECG, respiration, triaxial accelerometer) attached to a chest band that transmit sensor measurements wirelessly to a smart phone. Fine-grained measurement of physiology enables computation of physiological measures that are predictive of stress such as minute ventilation and heart rate variability. The project uses sophisticated machine learning methods to estimate stress levels, smoking episodes and conversation episodes from physiological measurements. In addition, accelerometers are used to identify changes in posture and physical activity, and GPS measurements on the mobile phone are used to identify location and transportation modality. Seventy-two smokers who want to quit will be recruited. Two weeks prior to quit date, they will wear AutoSense for 24h in the field. Behavioral treatment for abstinence will begin in their visit to the lab prior to quitting. They wil wear AutoSense for 72h in the field, beginning on the quit date. In addition to collecting continuous measurements across all sensors, smoking and lapse events will be self-reported. Saliva samples will be collected for cortisol assessment, in addition to physiological measures of stress. Participants will report back to the lab one week after the quit date to determine smoking/lapse status.
The aim of this project is to identify psychophysiological measures of stress and environmental cues that may predict early lapse. First, it will discover patterns in physiological and hormonal measures in response to withdrawal stress and acute stress that may predict lapse in the first week. Second, it will investigate the role of social and environmental context such as conversation behavior (e.g., duration and frequency of conversations), lifestyle factors (e.g., time spent in seated posture and in commuting), location (time spent at home vs. office), and indoor/outdoor status in mediating the relationship between stress and early lapse.

Public Health Relevance

By identifying psychophysiological and biobehavioral predictors of smoking lapse and by detecting smoking cues in the natural environment, this project will enable the development and evaluation of personalized cessation interventions that can be administered in real-time on smart phones, which could prove potent in preventing relapse. Given the adverse impact of smoking on smokers and non-smokers, such advancement will significantly improve the quality of life and reduce public health burden.

National Institute of Health (NIH)
National Institute on Drug Abuse (NIDA)
Research Project (R01)
Project #
Application #
Study Section
Special Emphasis Panel (ZRG1)
Program Officer
Grossman, Debra
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
University of Memphis
Biostatistics & Other Math Sci
Schools of Arts and Sciences
United States
Zip Code
Saleheen, Nazir; Chakraborty, Supriyo; Ali, Nasir et al. (2016) mSieve: Differential Behavioral Privacy in Time Series of Mobile Sensor Data. Proc ACM Int Conf Ubiquitous Comput 2016:706-717
Adams, Roy J; Saleheen, Nazir; Thomaz, Edison et al. (2016) Hierarchical Span-Based Conditional Random Fields for Labeling and Segmenting Events in Wearable Sensor Data Streams. Proc Int Conf Mach Learn 48:334-343
Chatterjee, Soujanya; Hovsepian, Karen; Sarker, Hillol et al. (2016) mCrave: Continuous Estimation of Craving During Smoking Cessation. Proc ACM Int Conf Ubiquitous Comput 2016:863-874
Hovsepian, Karen; al'Absi, Mustafa; Ertin, Emre et al. (2015) Stress: Towards a Gold Standard for Continuous Stress Assessment in the Mobile Environment. Proc ACM Int Conf Ubiquitous Comput 2015:493-504
Saleheen, Nazir; Ali, Amin Ahsan; Hossain, Syed Monowar et al. (2015) puffMarker: A Multi-Sensor Approach for Pinpointing the Timing of First Lapse in Smoking Cessation. Proc ACM Int Conf Ubiquitous Comput 2015:999-1010
Kennedy, Ashley P; Epstein, David H; Jobes, Michelle L et al. (2015) Continuous in-the-field measurement of heart rate: Correlates of drug use, craving, stress, and mood in polydrug users. Drug Alcohol Depend 151:159-66
Sharmin, Moushumi; Raij, Andrew; Epstien, David et al. (2015) Visualization of Time-Series Sensor Data to Inform the Design of Just-In-Time Adaptive Stress Interventions. Proc ACM Int Conf Ubiquitous Comput 2015:505-516
Rahman, Mahbubur; Bari, Rummana; Ali, Amin Ahsan et al. (2014) Are We There Yet? Feasibility of Continuous Stress Assessment via Wireless Physiological Sensors. ACM BCB 2014:479-488
Hossain, Syed Monowar; Ali, Amin Ahsan; Rahman, Mahbubur et al. (2014) Identifying Drug (Cocaine) Intake Events from Acute Physiological Response in the Presence of Free-living Physical Activity. IPSN 2014:71-82
Vhaduri, Sudip; Ali, Amin; Sharmin, Moushumi et al. (2014) Estimating Drivers' Stress from GPS Traces. Proc Int Conf Automot User Interfaces Interact Veh Appl (2014) 2014:

Showing the most recent 10 out of 12 publications