Research in Personal Health Informatics at Northeastern

The research projects below may not fully represent all active projects. Prospective students are encouraged to contact PHI faculty and inquire about specific projects and research interests. New projects are always in development.



Virtual Advisors for Physical Activity Promotion in Underserved Communities
PHI Faculty: Timothy Bickmore

The purpose of this project is to further develop a computer-animated exercise coach to promote physical activity among older English and
Spanish-speaking Latino adults, and to conduct a large-scale randomized trial at community centers in the San Francisco Bay Area.
Collaborator(s): Stanford University School of Medicine
Supported by: NIH/National Heart Lung and Blood Institute



Research Ethics and Safety Promoted by Embodied Conversational Technology (RESPECT)
PHI Faculty: Timothy Bickmore

This effort aims to develop conversational agents that help cancer patients with low health literacy through the oncology clinical trials process. One agent will help patients find trials they are eligible for. A second agent will automate the informed consent process. A final agent will assist patients during a clinical trial, by promoting adherence to study protocols, and detecting and reporting adverse events.
Collaborator(s): Boston Medical Center
Supported by: NIH/NCI
See more..


Preconception Care
PHI Faculty: Timothy Bickmore

The purpose of this project is to develop a relational agent to deliver preconception care information to African American women of childbearing age, and to evaluate this system in a randomized clinical trial with a national sample of 500 women.
Collaborator(s): Boston Medical Center
Supported by: NIH/NIMHD


An Always On Relational Agent for Social Support of Older Adults
PHI Faculty: Timothy Bickmore

The purpose of this project is the creation of intelligent virtual and robotic agents capable of providing social, instrumental, and social network support to older adults living alone. The resulting agents will be capable of continuous, always on, operation, with persistent social and relational goals, enabling them to proactively provide support. The agents will be evaluated in a series of longitudinal studies in which several virtual and robotic agents are installed in homes during the third and fourth years of the effort.
Collaborator(s): Worcester Polytechnic Institute
Supported by: NSF/HCC


A Heart Healthy Action Program for Puerto Rican Adults
PHI Faculty: Katherine Tucker, Carmen Sceppa, Timothy Bickmore

The goal of this project is to create an effective and sustainable multi-level heart healthy action program for Puerto Rican adults to reduce the risk of and the complication from cardio-vascular disease. We are developing a bi-lingual, culturally-tailored, conversational agent to provide a year-long intervention to promote heart health through exercise, diet, and stress reduction.
Supported by: NIH/NHLBI


Using Computer Agents to Provide Information and Support to Breastfeeding Mothers
PHI Faculty: Roger Edwards, Timothy Bickmore

The purpose of the project is to develop a relational agent that promotes breastfeeding among new mothers, intervening during the third trimester to boost motivation, bedside in the maternity ward to provide information on breastfeeding basics, and at home six months postnatally to promote the CDC-recommended six months of exclusive breastfeeding.
Collaborator(s): Melrose-Wakefield Hospital
Supported by: Hood Foundation

Optimizing Hospital Workflow and Quality through Patient Engagement
PHI Faculty: Timothy Bickmore, Harriet Fell, Stephen Intille

The Hospital Buddy is a computer agent that provides continual health counseling and companionship during a patient’s hospital stay via a bedside touchscreen computer equipped with sensors (RFID tags on staff, accelerometer on patient, acoustics) that can detect certain events occurring in the hospital room.
Collaborator(s): Boston Medical Center
Supported by: CIMIT

Reducing Chronic Pain Using Group Outpatient Visits and Relational Agents
PHI Faculty: Timothy Bickmore

The purpose of this project is to develop a relational agent to deliver stress reduction, exercise promotion, and nutrition information to participants with chronic pain and mild depression at Boston Medical Center, and to evaluate it in a randomized clinical trial.
Collaborator(s): Boston Medical Center
Supported by: PCORI

Cellphone Intervention Trial for You (CITY)
PHI Faculty: Stephen Intille

A five-year study to develop and evaluate (in a randomized clinical trial) sensor-enabled mobile phone technology to assist young adults aged 18-35 with long-term weight loss and weight management for two years.
Collaborator(s): Duke Medical School
Supported by: NIH/NHLBI
See more..

Encouraging GEI Activity Monitor Adoption: Demonstrating Device Equivalency
PHI Faculty: Stephen Intille

A project using custom-designed mechanical shakers and pattern recognition algorithms to demonstrate how phones can be used to produce output nearly equivalent to existing physical activity monitors.
Collaborator(s): Stanford Medical School
Supported by: NIH Genes and Environment Initiative

Development of a Time Use Intervention Using Mobile Phones to Promote Physical Activity in Youth
PHI Faculty: Stephen Intille

Several projects to explore the use of experience sampling on mobile phones for physical activity data gathering in children and adults.
Collaborator(s): University of Southern California Medical School
Supported by: Robert Wood Johnson Foundation

Development of Optimal Monitor Placement and Accelerometer Algorithms for Detecting the Activities of Adults and Children
PHI Faculty: Stephen Intille

Various projects to study the use of accelerometry-based motion monitoring (with monitors at a variety of locations) to improve a wearable, personal activity and energy expenditure detection in adults and children.
Collaborator(s): EveryFit, Inc., Stanford Medical School,
Supported by: NIH/NCI and NIH/NHLBI


Development of Longitudinal Home Activity Datasets as a Shared Resource
PHI Faculty: Stephen Intille

This project is developing portable sensor tools that can be used in typical homes to collect data for computer science and health research, as well as to generate shared datasets on home activity from actual homes to be used as a community resource to accelerate research.
Collaborator(s): MIT
Supported by: National Science Foundation
See more..


Enabling Population-Scale Physical Activity Measurement on Common Mobile Phones
PHI Faculty: Stephen Intille

A study with Stanford School of Medicine to create novel health monitoring tools for mobile phones. Includes a supplement to develop mobile context-sensitive ecological momentary assessment software for mobile phones (“Extensible Platform for Implementing Experience Sampling on Mobile Phones”)
Collaborator(s): Stanford Medical School
Supported by: NIH/NHLIB (NIH Genes and Environment Initiative)
See more..

End-User-Driven Training of Activity Recognition Algorithms
PHI Faculty: Stephen Intille

A project investigating the use of in-home context sensing, where end-users drive the algorithm training process, as applied towards proactive health care.
Supported by: Intel

Telemetric Assessment of Movement Stereotypy in Children with ASD
PHI Faculty: Matthew Goodwin, Stephen Intille

A study with the Groden Center, a school for autistic children, and the University of Rhode Island to explore the use of wireless accelerometers for automatic detection of autistic stereotypies.
Collaborator(s): Groden Center, University of Rhode Island
Supported by: National Alliance for Autism

Generating a Free, High-Quality Food Product Database using Games with a Purpose
PHI Faculty: Stephen Intille

The specific aim of this project is to use a “games with a purpose” approach as a proof-of-concept to construct a UPC and nutrient database.
Supported by: NIH/NCI

Using Mobile Phones to Reduce Missing Data in Youth Activity Monitoring Studies
PHI Faculty: Stephen Intille

The overall objective of this project is to develop new software for common mobile phones that can both reduce and explain missing data collected during objective and EMA activity monitoring studies with free living adolescents. This technology will supplement objective monitors already used today, with minimal additional cost.
Collaborator(s): University of Southern California Medical School
Supported by: Pending

Automatic Detection of Smoking Behavior
PHI Faculty: Stephen Intille

The goal of this pilot project is to develop and evaluate real-time pattern recognition algorithms that run on mobile phones and process data from miniature, wearable motion sensors (accelerometers) to detect patterns of smoking behaviors in free-living individuals.
Collaborator(s): M.D. Anderson Cancer Research Center
Supported by: M.D. Anderson Cancer Research Center

Maternal Stress and Children’s Obesity Risk (MATCH study)
PHI Faculty: Stephen Intille

The goal of this project is to determine whether levels of stress among working mothers are related to increased obesity risk in their children. It will use novel methods such as ecological momentary assessment to examine within-day mother-to-child stress processes that contribute to children’s long-term obesity risk in an accumulated manner over time. Working mothers and their 9 to 11 year-old children will participate in 6 semi-annual assessments waves across 3 years.
Collaborator(s): University of Southern California Medical School
Supported by: NIH/NHLBI

Lifestyle Modification and Social Support Via Text Messages
PHI Faculty: Stephen Intille

The goal of this project is to develop a completely new way of supporting lifestyle modification using just-in-time and context-sensitive text messaging.

Multimodal Computational Behavior Analysis
PHI Faculty: Matthew Goodwin

This project will define and explore a new research area we call Computational Behavior Science – integrated technologies for multimodal computational sensing and modeling to capture, measure, analyze, and understand human behaviors. Just as 20th century medical imaging technologies revolutionized internal medicine, we believe behavior imaging technologies will usher in a new era of quantitative understanding of behavior. Our motivating goal is to revolutionize the diagnosis and treatment of behavioral and developmental disorders. Our thesis is that emerging sensing and interpretation capabilities in vision, audition, and wearable computing technologies, when further developed and properly integrated, will transform this vision into reality. The need for this technology and its potential impact, both societal and scientific, are obvious and broad, ranging from the cognitive and brain sciences to the etiology, diagnosis, and treatment of developmental disorders such as autism. More specifically, we hope to: (1) Enable widespread screening of autism by allowing non-experts to easily collect high-quality behavioral data and perform an initial assessment of risk status; (2) Improve behavioral therapy through increased availability and improved quality, by making it easier to track the progress of an intervention and follow guidelines for maximizing learning progress; and (3) Enable longitudinal analysis of a child’s development based on quantitative behavioral data, using new tools for visualization.
Collaborator(s): Georgia Tech, MIT Media Lab, Boston University, University of Southern California, Carnegie Mellon University, Emory University, University of Illinois at Urbana Champaign
Supported by: National Science Foundation, Expeditions Award

Innovative Technology for Understanding & Assisting Autism Spectrum Disorders
PHI Faculty: Matthew Goodwin

This project will extend work in the areas of wireless autonomic sensing, wireless motor movement sensing, and non-invasive audio and video recording to service the autism community. These three core technology platforms carry the potential to revolutionize behavior science research in ASD, and developmental science more broadly, by enabling unobtrusive, longitudinal, and ecologically valid assessments of biology, behavior, and environmental context. In addition to the scientific merits of conducting research in this way, there are abundant opportunities to use the technology we are developing in a clinical capacity to help individuals with ASD, their families, and teachers to better understand, support, and increase quality of life outcomes. The vast majority of autism research is dedicated to epidemiology, genetics, and neuroscience. While critical for understanding prevalence, etiology, and underlying mechanisms of behavior, very little of this research translates into practical knowledge, tools, or techniques that can beneficially impact the lives of the growing number of people who currently have ASD. This project aims to redress this imbalance (while contributing scientific data to epidemiologists, geneticists, and neuroscientists) by extending and adapting the technologies we develop to be used by other autism specialists, persons with ASD, their families, and support networks.
Collaborator(s): MIT Media Lab, The Groden Center, Inc., The Center for Discovery, The Lurie Center/MGH, Youthcare/MGH
Supported by: Nancy Lurie Marks Family Foundation

The Speechome Project: Dense Corpora of ASD Children’s Speech and Interactions May Predict Language Development and Outcome
PHI Faculty: Matthew Goodwin

This project is a supplement to Dr. Naigels’s NIH R01 study “Language Development and Outcome in Children with Autism,” that aims to: (1) Replicate previous findings in ASD children of language comprehension preceding production and of the use/nonuse of specific linguistic principles; (2) Investigate the relationships between ASD children’s early language development and their later language/cognitive outcome; and (3) Investigate how more detailed measures of on-line efficiency in language comprehension might predict ASD children’s individual variation. The overall method involves a longitudinal collection of language comprehension and production data from children with ASD and typically developing children at four-month intervals for a total of 6 visits, with a comprehensive follow-up collected two years after the last visit. This supplement contributes to and extends all three specific aims by providing an ecologically-valid, densely sampled, and extremely efficiently analyzed audio-visual corpus of each child’s speech and home environment. Methodologically, data collection of the supplement will be ‘piggy-backed’ onto the data collection procedures of a subset of the current sample of participants. In particular, the supplement will enable the most valid assessments of children’s levels of speech production thus far, by including daily 2-3 hour samples in the home environment over a period of 2-12 months. Thus, we will be able to ascertain each child’s speech sophistication on a variety of lexical, grammatical, and pragmatic levels, capture the shape of his/her developmental change at these levels, and compare these with his/her levels of language comprehension. This project is in line with recent studies of young typically developing children’s motor and language development, which have revealed that such ‘dense’ sampling vastly increases sensitivity to the occurrence and non-occurrence of words/motor behaviors, thus rendering more accurately the patterns of development involved in their use.
Collaborator(s): MIT Media Lab, University of Connecticut
Supported by: NIH-DCD Supplement, Nancy Lurie Marks Family Foundation

Innovative Assessments of Top-down Processing in Persons with Autism Across a Range of Functioning Levels
PHI Faculty: Matthew Goodwin

The goal of this project is to undertake innovative, systematic tests of top-down processing in autism. In pursuit of this goal, an important consideration for us is to design our studies in such a way that we can enlist participation from individuals on the autism spectrum who have cognitive impairments and limited verbal abilities. This will have the twin benefits of mitigating subject-selection biases in extant study results on the one hand, and developing practical experimental tools that can bring this hitherto neglected group into the fold of scientific inquiry and understanding. We propose to use RISE, an image presentation paradigm developed a few years ago in the Sinha Lab, to sensitively assess top-down influences in neurotypical and autistic participants. RISE is a computationally rigorous and systematic way to assess the limits of recognition processes and derive precise estimates of an individual’s recognition abilities. It assesses how a person’s response changes as one moves along a continuous trajectory passing through a point of interest in a multidimensional image space. We will investigate whether the amount of hysteresis evident in neurotypicals is different from that seen in individuals with autism. For neurotypicals, we can simply use their verbal responses to reliably estimate onset and offset points. However, in working with individuals on the autism spectrum, especially those who lack communication abilities, reliance on verbal responses is problematic. A given participant might be hesitant about speaking or be entirely non-verbal. Also, traditional methods of brain-imaging are too obtrusive to get compliance from persons with autism who have cognitive impairments. To overcome this difficulty, we propose an innovative battery of unobtrusive assessment techniques that characterize an individual’s involuntary responses to a stimulus. This battery will include eye-tracking, pupillary dilation, and wireless recording of electrodermal activity.
Collaborator(s): MIT Media Lab, Brain Cognitive Science, MIT
Supported by: Simons Foundation

Wearable Wireless Toolkit for Measurement and Communication of Autonomic Nervous System in Autism
PHI Faculty: Matthew Goodwin

People with ASD who have written about their experiences almost always describe immense stress and anxiety. While many scientists have tried to characterize stress and other Autonomic Nervous System (ANS) responses associated with ASD, traditional measurements have been limited to single observations in an artificial laboratory setting, and to average characterizations of groups that do not address the variability or other patterns in an individual’s ANS responsivity throughout natural daily activities. The key problem that has held back progress with these studies is that existing measurement tools have been impossible to use in a continuous unobtrusive way outside the laboratory and with form factors that fall off and collect too much noise under normal active movement conditions. The current research utilizes state-of-the-art knowledge in technology, especially in wearable sensors and wireless technology, to construct a comfortable, low-cost wireless toolkit that makes it possible for people with ASD and their caregivers to continuously monitor and communicate autonomic arousal in daily life, including activity at home, school, and in community settings. To the extent that such technology is adapted for use by people who have difficulties communicating how they feel, there arise abundant opportunities for individuals to learn about and communicate their internal states with others. For instance, an individual with minimal verbal abilities might be able to communicate a situation causing stressful overload or anxiety which might not be obvious to observers– perhaps an internal pain or temporally distant external stimulus that is overwhelming. Of particular focus in the current study, there is also reason to believe that ANS patterns may provide early indications not only of stress-related events, but also of life-threatening events such as impending seizures. The having of repeated seizures (or epilepsy) is a condition that is conservatively estimated to occur in 25% of people with ASD. The technology we are building will enable measurement of ANS and physical responsivity that may possibly provide early warning indications of seizures and alert caregivers to make appropriate accommodations at the precise moment when they are needed to prevent or minimize the impact of a seizure.
Collaborator(s): MIT Media Lab, The Groden Center, Inc., Boston Children’s Hospital
Supported by: Nancy Lurie Marks Family Foundation

Assessing and Communicating Movement Stereotypy and Arousal Telemetrically in Individuals with Autism Spectrum DisorderPHI Faculty: Matthew Goodwin

Stereotypical motor movements are one of the most common and least understood behaviors occurring in individuals with ASD. Stereotypical motor movements are complex and thought to serve a multiplicity of functions. While no one theory has obtained overwhelming support, there is evidence for biological, operant, and homeostatic interpretations. Of particular importance to the current project, a small number of studies support the notion that there is a functional relationship between movement stereotypy and arousal in individuals with ASD, such that changes in autonomic activity either precede or are a consequence of engaging in stereotypical motor movements. Thus, it appears to be the case for some individuals that stereotypical movements are adaptively employed to help regulate stress, which in turn may help regulate attention, emotion, and social behaviors. Unfortunately, however, it is difficult to generalize these findings since previous studies fail to report reliability statistics that demonstrate accurate identification of movement stereotypy start and end times, and use autonomic monitors that are obtrusive and thus only suitable for short-term measurement in laboratory settings. The current investigation further explores the relationship between movement stereotypy and autonomic activity in persons with ASD by combining state-of-the-art ambulatory heart rate monitors to objectively assess arousal across settings and wireless, wearable motion sensors and pattern recognition software that can automatically and reliably detect stereotypical motor movements in individuals with ASD in real-time. Obtaining detailed and accurate information on the occurrence, type of movement (i.e., topography), frequency, duration, and setting events associated with movement stereotypy is critical to understanding this behavior. Moreover, assessing and communicating stereotypical movements and arousal telemetrically may facilitate more precise intervention efforts before they are entrenched in an individual’s repertoire. Stereotypical behaviors are often evoked under periods of transition and stress, thus creating greater difficulty for caregivers and educators to monitor and redirect an individual from excessive stereotypic engagement. The measures being developed in this study may allow preventative adjustment in stress and adaptive functioning not currently available in response-oriented therapies.
Collaborator(s): MIT Media Lab, The Groden Center, Inc.
Supported by: Nancy Lurie Marks Family Foundation

Social-Emotional Technologies for Autism Spectrum Disorders
Spectrum DisorderPHI Faculty: Matthew Goodwin

This project is developing and evaluating wearable social-emotional technology that helps individuals with high-functioning autism and/or Asperger syndrome acquire an affinity for the social domain and improve their overall social abilities. We are creating the first wearable camera system capable of perceiving and visualizing social-emotional information in real-time human interaction. Using a small wearable camera and video-pattern analysis algorithms, the system analyzes video of the wearer and interaction partner and tags it at multiple granularities (facial actions, communicative facial/head gestures). The wearable system aims to: (1) facilitate learning and systemizing of social-emotional cues; (2) promote self-reflection and perspective-taking; (3) allow wearers to study subtle nonverbal cues and share experiences with peers, family members, and caregivers; and (4) contribute new computational models and theories of social-emotional intelligence in machines. A clinical study will compare the efficacy of the wearable system to current gold standard interventions for ASD. Our human-centered, participatory approach to the co-design and use of technology draws on the experiences of individuals with ASD and their solutions to systematizing social interactions, thereby empowering them to enhance their relationships while participating in the development of next-generation social-emotional intelligent technologies.
Collaborator(s): MIT Media Lab, The Groden Center, Inc.
Supported by: National Science Foundation



VocalID: An Adaptive Speech Synthesizer that Conveys User Identity
PHI Faculty: Rupal Patel

Users of assistive communication technologies rely on generic and robotic sounding synthetic voices that impede social integration and technology ownership. This project aims to develop personalized synthetic voices for individuals with severe speech disorders by leveraging the users’ residual vocal abilities. Voice morphing techniques are used to project the target user’s speaker identity cues onto a healthy talker database. The resultant speech sounds like the target user in vocal identity yet is clear and intelligible.
Collaborator(s): A.I. duPont Hospital for Children
Supported by: National Science Foundation
See more..



ReadN’Karaoke: Visualing Speech Prosody to Improve Oral Reading Expressiveness
PHI Faculty: Rupal Patel

Beginning readers struggle to read aloud with appropriate intonation and rhythm, which in turn impacts reading comprehension. Written text does not provide sufficient cues for how to parse sentences and where to place emphasis.
We have developed software that provides explicit visual cues about pitch, loudness and duration that are integrated with text to improve oral reading expressivity. The software has implications for improving reading fluency and comprehension in typically developing children and second language learners as well as therapeutic value for children on the autism spectrum who have particular difficulty with speech melody.
Supported by: National Science Foundation
See more..

RSVP-iconCHAT: An icon-based brain computer communication interface
PHI Faculty: Rupal Patel, Deniz Erdogmus

Many individuals with severe communication disorders also have concomitant physical impairments that preclude the use of conventional modes of communication. We have developed an icon-based interface in which concepts are displayed in rapid serial visual presentation (RSVP) within semantic frames to enable individuals to construct messages via a single input key. The input modality may be a keystroke or a brain signal sensed through electrodes placed on the surface of the scalp. The system has direct implications for users with severe speech and mobility impairments and potential application as a language translation/mediation tool on mobile platforms.
Supported by: National Science Foundation
See more..

Motivation Game Mechanics for Behavior Change in Health Based Games
PHI Faculty: Magy Seif El-Nasr

Recently, researchers have started exploring the potential utility of social networks and games in inducing behavior change, and potentially breaking some habits. In order to use the power of games and social networks, researchers need to understand the systems that make games engaging and appealing. This project explores the use of a framework based on psychology of motivation called Player Experience of Need Satisfaction, developed by Rigby and Ryan. In particular, we explore the the goal of influencing behavior change via the development of engagement models integrating real-time behavior tracking, selective information visualization coupled with social structure facilitating cooperation, and competition and casual games. The model is currently being deployed in an industry product, which we will be using to derive behavioral data to influence and revise the design.
Industry Partner: Ignite Play, Vancouver BC Canada
Supported by: GRAND, NCE, NSERC

A Platform for Automated Acoustic Analysis for Assistive Technology
PHI Faculty: Harriet Fell

The use of speech production data has been limited by a steep learning curve and the need for laborious hand measurement. We are building a tool set that provides summary statistics for measures designed by clinicians to screen, diagnose or provide training to assistive technology users. This will be achieved by extending an existing shareware software platform with “plug-ins” that perform specific measures and report results to the user. The common underlying basis for this tool set is a Stevens’ paradigm of landmarks, points in an utterance around which information about articulatory events can be extracted.
Collaborator(s): University of Cincinnati, Boston University, Speech Technology and Applied Research
Supported by: NIH
See more..

Automated Tools for Identifying Syllabic Landmark Clusters that Reflect Changes in Articulation
PHI Faculty: Harriet Fell

We have developed a set of software tools to detect articulatory changes in the production of syllabic units based on acoustic landmark detection and classification. Results from the application of this automatic analysis system to studies of Parkinson’s Disease and Sleep Deprivation show the ability to detect subtle change. We are making these tools available as add-ons to systems such as Wavesurfer and R.
Collaborator(s): University of Cincinnati, Speech Technology and Applied Research
Supported by: NIH
See more..

Innovative Speech Articulation Tools for Neuroscience Research
PHI Faculty: Harriet Fell

Changes in health status and speaking condition have been shown to affect the articulation of speech. Scientific evaluation of these changes has been handicapped by lack of access to robust, user-friendly software tools for fast automatic measurement of speech acoustics. We are providing tools for objective assessment of changes in speech production that might result from assistive technology interventions.
Collaborator(s): University of Cincinnati, Speech Technology and Applied Research
Supported by: NIH

RSVP Keyboard
PHI Faculty: Deniz Erdogmus

We are developing a brain interface for locked in persons to type. Rapid serial visual presentation paradigm is used to induce event-related potentials, in particular the novelty response P300. Our prototype system can perform at over 95% symbol selection accuracy and we are currently working on increasing speed of typing as well as accuracy.
Collaborator(s): Oregon Health and Science University (OHSU)
Supported by: NIH

Human-in-the-loop Cyber-Physical Systems
PHI Faculty: Deniz Erdogmus

In this project, we will be developing brain controlled robot applications running on embedded platforms. Robotic applications will target locked-in persons and increasing their quality of life and independence.
Collaborator(s): University of Cincinnati, Speech Technology and Applied Research, Worcester Polytechnic Institute
Supported by: NIH

Interactive Computing for Family-Based Physical Activity Promotion
PHI Faculty: Andrea Parker

Over 25% of preschoolers in the United States are overweight or obese and rates are even higher among children from low-income, ethnic minority families. Despite the high prevalence of overweight and obesity among young children, prevention efforts are just beginning to focus on this age group and there has been a national call to expand these efforts. Sedentary living is one major contributing factor to childhood obesity; as such, interventions that help young children increase their physical activity levels are needed. Involving parents in these interventions is critical, given parents’ influence over children’s physical activity environments and the powerful impact they have as role models. In this research, we are conducting formative research with families in low-income Boston neighborhoods to identify design requirements for interactive applications that help increase physical activity within the family unit, that is, in both parents and young children. Through this work we are translating these requirements into a functional prototype that leverages activity monitoring devices and novel visualizations of that physical activity data to encourage increased activity in the parent/caregiver-child context.

Technology-Supported Youth Advocacy
PHI Faculty: Andrea Parker

Focusing on low-socioeconomic status populations, we are examining how interactive computing applications can empower youth to advocate healthy behaviors within their social networks and communities. Involving youth in such participatory action is critical for their personal development and wellbeing, but also for society: youth bring unique creativity, insights, and facility with using technology such as social networking applications (SNAs), platforms through which large-scale advocacy and social organizing can take place. Innovations in SNA design have the potential to revolutionize youth participation in societal change by giving teens an authentic platform on which their voice can be heard, a broader audience to share that voice with, and facilitating collective action. This work seeks to help youth identify the strengths and barriers to wellness in their neighborhoods (e.g., the prevalence of fast food) and how innovative tools might help community residents to overcome these barriers. Building upon social ecological theories, we are conducting research with teens to identify the benefits of, and challenges to, using existing SNAs for health advocacy and design guidelines for future tools. The long-term goal of this work is to build and evaluate this next generation of systems that enable youth to take an active role in encouraging change within their communities.

Interactive Tools to Support Healthy Eating in Children
PHI Faculty: Andrea Parker

This research explores how technology can support health promotion in children by making nutrition education an enjoyable and engaging experience. Using a participatory design methodology, we are conducting research with children to identify design concepts and guidelines for such interactive tools.

Relational Agent Screening and Brief Counseling for Alcohol Abuse in the Veterean’s Administration
PHI Faculty: Timothy Bickmore

The purpose of this project is to develop and evaluate a relational agent that will screen veterans for alcohol and substance abuse in the outpatient setting, and provide brief counseling if problems are identified.
Collaborator(s): Boston VAMC
Supported by: VA

Collecting Family Health History Using Relational Agents
PHI Faculty: Timothy Bickmore

The purpose this project is to design a relational agent that takes
patient family medical history in preparation for a clinical consultation.
Collaborator(s): Boston University School of Public Health

Depression Counseling by Relational Agents for Patients with CHF or COPD following Hospitalization
PHI Faculty: Timothy Bickmore

The purpose of this project is to develop and evaluate a relational agent that treats depression in patients with Congestive Heart Failure (CHF) or Chronic Obstructive Pulmonary Disease (COPD) after being discharged from the hospital.
Collaborator(s): Boston Medical Center, Harvard Medical School/Mass General

Brain-Computer Interface for Signaling Changes in Psychological States
PHI Faculty: Lisa Barrett, Deniz Erdogmus, Stephen Intille, Matthew Goodwin

The major goal of this project is establish the feasibility of using fNIRS to detect major changes in emotional state and to use a mobile system for context-sensitive self-report on that state change. Collaborator(s): Dana Brooks (NEU)
Supported by: Northeastern Tier 1 Grant Program

Modeling Temporally-Dense Microinteractions to Promote Health Behavior Change
PHI Faculty: Stephen Intille, Matthew Goodwin, Karen Quigley

The purpose of this project is to study how new microinteraction devices such as Google Glass can be used for measuring health behaviors and behavior change.
Collaborator(s): Donna Spruijt-Metz (USC)
Supported by: Google, Inc.

Interactive Speech Assessment for Children
PHI Faculty: Rupal Patel

A contextual, connected speech sample can provide essential information when assessing childhood motor speech disorders; however, a standard tool specifically tailored for this population has not been generally implemented. To address this need, we designed a child-friendly picture scene with visual tokens that range in speech motor difficulty and familiarity to inform the diagnosis and differentiation of childhood motor speech disorders. We are now developing an interactive version of the scene that can be used to record spoken samples, reinforce spoken descriptions, and perform automated analyses.