News:
August 2017 and later
Summer, 2022
Congratulations on paper acceptances
Congratulations to Rithika and Binod on paper acceptances and Aditya on his UbiComp presentation.
December, 2021
Congratulations to Dr. Aditya Ponnada!
Congratulations to Dr. Aditya Ponnada (who will be known to us as "Dr. Aditya"), who received his degree at the end of December!
August 31, 2021
Congratulations to Dr. Qu Tang!
Congratulations to Dr. Qu Tang (who will be known to us as "Dr. Qu"), who received his degree at the end of August!
August 19, 2021
To Sit Less and Promote Healthy Activity, AI May be Key
A short article about mHealth collaborator Diego Arguello's work was in the NU News.
June 30, 2021
mHealth Group looking for new PhD students for fall 2022
The mHealth group is looking for new PhD students to join the research group in the fall of 2022. Applications for either the personal health informatics or the computer science doctoral programs are due December 15, 2021 See the Get Involved page for more information.
September, 2020
New grant with collaborators at Temple to study intervention for wheelchair users with spinal cord injuries
The mHealth group has been awarded a new three-year grant with Temple University to conduct participatory design research with individuals with spinal cord injuries (SCIs) and their friends and/or family members. The research involves iteratively developing and testing prototype just-in-time adaptive intervention (JITAI) systems designed to be sensitive to the individual and interpersonal needs of individuals with SCI and their support networks.
August, 2020
New grant to study "Accelerating the development of novel methods to measure 24-hr physical behavior"
Accurate measurement of physical activity, sedentary behavior, and sleep is necessary to support new science on behavior-related dose-response for major health conditions. In a new grant awarded to Prof. Intille and collaborator Prof. Dinesh John, we are developing novel methods to measure 24-hour physical behavior, as well as procedures via which those methods can be compared to others. We aim to help the research community to converge on methods that use the devices to accurately measure physical activity type and intensity, sedentary behavior and posture, and sleep in adults. The research involves collection of intensive longitudinal data of people performing everyday activities will wearing sensors and making such data available to the research community to drive new algorithm development.
May, 2020
mHealth Group looking for a postdoc
The mHealth group is looking for a postdoc interested in working on physical activity, sedentary behavior, and sleep activity recognition algorithms and systems. Check back here for more information soon...
October, 2018
mHealth Group awarded a new grant to study Smart and Connected Churches for Promoting Health in Disadvantaged Populations
The mHealth group, with collaborators at Northeastern and Boston Medical Center has been awarded a new four-year NSF grant to study the use of mobile technology in church communities to promote health.
The goal of this project is to use virtual conversational agents, sensor-enabled mobile technologies, and AI in a research infrastructure designed to assist church congregations with providing health and wellness support to their members. One key goal is to create technical systems that adapt to individual behaviors.
The goal of this project is to use virtual conversational agents, sensor-enabled mobile technologies, and AI in a research infrastructure designed to assist church congregations with providing health and wellness support to their members. One key goal is to create technical systems that adapt to individual behaviors.
August, 2018
mHealth Group awarded a new grant to study Microtemporal Processes Underlying Health Behavior Adoption and Maintenance
The mHealth group, with collaborators at the University of Southern California, has been awarded a new four-year NIH grant to study habits using new smartwatch-based measurement technologies.
Abstract: Emerging adulthood (ages 18-24 years) is marked by substantial weight gain, leading to increased lifetime risks of cancer and other chronic diseases. Engaging in sufficient levels of physical activity and sleep, and limiting sedentary time are important contributors to the prevention of weight gain. However, engaging in these healthy behaviors peaks during the childhood and adolescent years, and steeply deteriorates into emerging adulthood. Interventions promoting physical activity, reduced sedentary time, and sufficient sleep typically focus on the adoption of these behaviors. Yet, when these interventions are successful, new patterns of behavior are not maintained and typically regress back to baseline levels. Traditional health behavior theories provide limited guidance regarding factors underlying behavior maintenance. To address this gap, our work suggests that dual-process models of decision-making and behavior can shed light on differences in the mechanisms underlying adoption versus maintenance. Reflective processes (e.g., efficacy, deliberations, self-control) may be activated to a greater extent during behavior adoption. In contrast, reactive processes (e.g., contextual cues, automaticity, habits) may play a greater role in behavior maintenance. However, reactive processes are difficult to measure using retrospective methods because they can unfold on a micro-timescale (i.e., change across minutes or hours). To solve this problem, we propose to use real-time mobile technologies to collect intensive longitudinal data examining differences in the micro-temporal processes underlying the adoption and maintenance of physical activity, low sedentary time, and sufficient sleep duration. We will conduct a prospective within- subject case-crossover observational study across a 12-month period. Ethnically-diverse, emerging adults (ages 18-24, N=300) will be recruited from the Happiness and Health Cohort (R01DA033296). We will conduct intermittent self-report (i.e., ecological momentary assessment) of reflective variables; and continuous, sensor- based passive monitoring of reactive variables (e.g., location, social proximity, voice/text communication) and behaviors (i.e., physical activity, sedentary time, sleep) using smartwatches and smartphones. These data will be used to predict within-subject variation (withindays, between-days) in the likelihood of behavior “episodes” (e.g., ≥10 min of physical activity, ≥120 min sedentary time, ≥7 hr sleep) and “lapses” (i.e., failure to achieve recommended levels ≥7 days). The specific aims are to (1) idiographically use machine learning to identify person-specific combinations of time-varying reflective and reactive factors that predict behavior episodes and lapse; and (2) nomothetically determine whether there are general, group-level patterns of time-varying predictors, and whether those patterns predict successful behavior maintenance outcomes. The data and methods from this project will contribute to the U01/U24 Intensive Longitudinal Behavior Initiative’s collective goal to build more predictive health behavior theories that specify targets for personalized interventions.
Abstract: Emerging adulthood (ages 18-24 years) is marked by substantial weight gain, leading to increased lifetime risks of cancer and other chronic diseases. Engaging in sufficient levels of physical activity and sleep, and limiting sedentary time are important contributors to the prevention of weight gain. However, engaging in these healthy behaviors peaks during the childhood and adolescent years, and steeply deteriorates into emerging adulthood. Interventions promoting physical activity, reduced sedentary time, and sufficient sleep typically focus on the adoption of these behaviors. Yet, when these interventions are successful, new patterns of behavior are not maintained and typically regress back to baseline levels. Traditional health behavior theories provide limited guidance regarding factors underlying behavior maintenance. To address this gap, our work suggests that dual-process models of decision-making and behavior can shed light on differences in the mechanisms underlying adoption versus maintenance. Reflective processes (e.g., efficacy, deliberations, self-control) may be activated to a greater extent during behavior adoption. In contrast, reactive processes (e.g., contextual cues, automaticity, habits) may play a greater role in behavior maintenance. However, reactive processes are difficult to measure using retrospective methods because they can unfold on a micro-timescale (i.e., change across minutes or hours). To solve this problem, we propose to use real-time mobile technologies to collect intensive longitudinal data examining differences in the micro-temporal processes underlying the adoption and maintenance of physical activity, low sedentary time, and sufficient sleep duration. We will conduct a prospective within- subject case-crossover observational study across a 12-month period. Ethnically-diverse, emerging adults (ages 18-24, N=300) will be recruited from the Happiness and Health Cohort (R01DA033296). We will conduct intermittent self-report (i.e., ecological momentary assessment) of reflective variables; and continuous, sensor- based passive monitoring of reactive variables (e.g., location, social proximity, voice/text communication) and behaviors (i.e., physical activity, sedentary time, sleep) using smartwatches and smartphones. These data will be used to predict within-subject variation (withindays, between-days) in the likelihood of behavior “episodes” (e.g., ≥10 min of physical activity, ≥120 min sedentary time, ≥7 hr sleep) and “lapses” (i.e., failure to achieve recommended levels ≥7 days). The specific aims are to (1) idiographically use machine learning to identify person-specific combinations of time-varying reflective and reactive factors that predict behavior episodes and lapse; and (2) nomothetically determine whether there are general, group-level patterns of time-varying predictors, and whether those patterns predict successful behavior maintenance outcomes. The data and methods from this project will contribute to the U01/U24 Intensive Longitudinal Behavior Initiative’s collective goal to build more predictive health behavior theories that specify targets for personalized interventions.
Summer, 2018
Ph.D. student Binod Thapa Chhetry works at Philips
Binod Thapa Chhetry accepted an internship to work for the summer at Philips Medical in Cambridge, MA to work on applications of machine learning to healthcare.
December, 2017
Dr. Intille invited to be an expert advisor for the "Using Technology to Prevent Childhood Obesity" federal challenge
In 2018, Dr. Intille will serve as an expert advisor for the "Using Technology to Prevent Childhood Obesity" federal challenge, hosted by the Health Resource Service Administration’s (HRSA’s) Maternal and Child Health Bureau (MCHB). The purpose of this challenge is to support the development of low-cost, scalable technology-based innovations to promote healthy weight for low-income children and families in the socio-cultural and environmental contexts of their communities.
September 14, 2017
Dr. Intille's paper receives a 10-year impact award
Dr. Intille's paper, "Real-Time Recognition of Physical Activities and Their Intensities Using Wireless Accelerometers and a Heart Rate Monitor," published in 2007 in the International Symposium on Wearable Computers received the conference's 10-Year Impact Award. Congratulations to authors Emmanuel Munguia Tapia, Stephen Intille, William Haskell, Kent Larson, Julie Wright, Abby King, and Robert Friedman.
September 13, 2017
A decade with smartphones
On Tuesday, Apple unveiled its newest generation of products at an event celebrating a decade since the first iPhone was released. Though it wasn’t the first smartphone on the market, its commercial success—and the subsequent saturation of smartphones across the globe—have changed society in massive ways, said two Northeastern professors.
Dr. Intille discussed the impact of smartphones on our everyday lives in the Northeastern News. Read more...
September 13, 2017
Aditya Ponnada presents at Ubicomp
Aditya Ponnada presented his IMWUT paper, “Microinteraction ecological momentary assessment response rates: Effect of microinteractions or the smartwatch?" at Ubicomp 2017 in Maui, Hawaii in September. The authors are Aditya Ponnada, Caitlin Haynes, Dharam Maniar, Justin Manjourides, and Stephen Intille.
August, 2017
mHealth group selected to process NHANES accelerometer data
The NHANES dataset is heavily used in behavioral science and health research.
The mHealth group was awarded a contract to use Amazon Web Services to process 18,000+ accelerometer datasets from the National Health and Nutrition Examination Survey (NHANES). This is work with Prof. Dinesh John.
Select older news
We overhauled our group website and have added just a few old news items here because we wanted to spend more time on research!
2014
mHealth Group receives two 10-Year Impact Awards
Two of Dr. Intille's papers received the Ubicomp/Pervasive 10-Year Impact Award in 2014 (for papers published in 2004). The papers were "Activity Recognition from User-Annotated Acceleration Data" (with co-author Ling Bao, an MS student at the time) and "Activity Recognition in the Home Using Simple and Ubiquitous Sensors" (with co-Authors Emmanuel Munguia Tapia, a PhD student at the time, and Kent Larson).
