Crowdsourcing technologies for improved public health surveillance
Rumi Chunara
Boston Children’s Hospital and Research Fellow at Harvard Medical School, HealthMap Group

Abstract: Global patterns of disease burden are constantly shifting. Multiple drivers, such as the shift of populations to urban centers, increased mobility and evolving human-animal interactions favor the emergence of new genetic variants of pathogens as well as increased burden of chronic diseases in low-, middle- and high-income countries. At the same time, traditional public health capacity has remained stagnant, suffering from limits resulting from latency, high cost, inherent contributor biases and imprecise resolution. A critical barrier in disease epidemiology and control has been a lack of rapid, quantitative intelligence on disease dynamics from the point of care. As an engineer working in epidemiology, the goal of my research is to address these gaps by building tools to obtain information directly from individuals, including Internet-based and point-of-care sensors using mobile phones, smart-messaging service (SMS), smart phone applications, portable bio-molecular diagnostics and rapid genotyping. Amongst other challenges, crowdsourced data from these sensors is generally unstructured, unvalidated and noisy. In this talk I will discuss how we approach these challenges in a variety of settings and diseases; malaria surveillance in India, cholera in Haiti and obesity in the United States, to obtain better spatial and temporal insight into disease dynamics.

Transforming Health Care with Technology: Model-Based Approaches
Misha Pavel
Program Director, Smart Health and Wellbeing, Computer and Information Science and
Engineering, National Science Foundation
Professor, Department of Biomedical Engineering and Department of Computer Science
and Electrical Engineering, Oregon Health and Science University

New Technology Approaches to Creating Scalable Health Interventions for a Changing World
Holly Jimison
Technology Advisor, National Institutes of Health, Office of the Director, Office of
Behavioral & Social Science Research
Associate Professor, Medical Informatics & Clinical Epidemiology, Oregon Health Science
University

Use of Pre-recorded Video and Audio to Create an Engaging Interactive Media Program for Mental Health
James Cartreine, PhD
Research and Clinical Psychologist
Program in Behavioral Informatics and eHealth
Department of Psychiatry, Brigham and Women’s Hospital
Instructor in Psychiatry, Harvard Medical School

Abstract: A 10-year program of research has been undertaken for NASA to develop and evaluate a suite of interactive media programs to help astronauts prevent, assess, and treat their own psychosocial problems. One of the Virtual Space Station tools that is now being transitioned for use by the public helps persons improve their mood and enjoyment of life, based on problem-solving therapy for depression. Through rich media (branching video and audio clips tailored to users’ inputs), a conversation is approximated between the user and a competent, caring therapist. The intent is to make the program feel more like interacting with a person than with a computer. Research studies have indicated that users “bond” with the on-camera therapist similar to live therapy. This talk will provide a rationale for when to use and not to use video, audio, graphics and text in a behavioral health-oriented computer program, as well as the process of storyboarding, scriptwriting, and media production that went into the production. A demonstration of the Virtual Space Station and its components will be provided.

RAGE-Control: Regulate and Gain Emotional Control
Jason Kahn
Boston Children’s Hospital and Harvard Medical School

Abstract: Many young children have clinical levels of anger and aggression, which has negative consequences with families, peers, and academics. Historically, these children have had to rely on psychotherapy or psychopharmacological medications to treat these symptoms. Both of these approaches can be effective, but have considerable limitations. Adapting both recent findings in neuroscience and constructivist principles, we have been developing video games to help these children by using active biofeedback, which requires both simultaneous attention and relaxation. In pilot studies in both outpatient and inpatient settings, we have found that using our video game leads to reduced symptoms of anger in children. We have taken these findings and started developing toys that use the principles.

Applying information technology to individualize health care
Tim Bunnell
Director, Nemours Bioinformatics Core Facility
Nemours/Alfred I. duPont Hospital for Children

Abstract: Three examples will be presented on how Nemours is leveraging information technology to reach patients (and sometimes providers) as individuals. Our Obesity Cluster project seeks to provide obese teens with a personal health coach, messaging, and feedback to help them pursue healthy lifestyle goals. At the same time, their primary care physicians are provided tools to track how well their management of obese patients (both in aggregate and individually) is succeeding to improve the health of their patients. A second project is exploring the use of mobile devices to facilitate the conduct of a clinical trial for asthma medication. The approach affords immediate data entry by patients, personalized reminders, and virtual visits with clinical staff and is expected to improved adherence to the study protocol, more timely detection of possible adverse events, and make participation feasible for patients who cannot easily make frequent in-person visits to a clinic. Finally, an iPad application will be described that is designed to assist in the delivery of Auditory-Verbal (AV) therapy for pediatric cochlear implant and hearing aid patients. This application will employ utterance verification technology that has been under development in the Nemours Speech Research Laboratory for several years now.

From MIMIC to Sana: Learning systems, global eHealth, quality improvement and design thinking
Leo Celi
MIT

Abstract: The failure to store and analyze the vast amount of data generated on a daily basis is a key hurdle in advancing the practice of medicine. An equally necessary and important goal involves creating a culture that reacts to and translates the findings of such “big data” into a better system of care. Together, these changes will close a feedback control loop that can help provide a more standardized, yet more personalized care. I will describe an interdisciplinary team-led, data-fueled, learning system that we have built around MIMIC, a public clinical database of patients admitted to the intensive care units at Beth Israel Deaconess Medical Center. I will also discuss the evolution of Sana, a project hosted at the Computer Science and Artificial Intelligence Laboratory at MIT, from an mHealth solution to an organization that promotes capacity building and design thinking in the development of health information systems to improve the quality of care in resource-poor settings.

Role of Patient-facing Technologies in the Era of Health Reform
David Ahern
Brigham and Women’s Hospital

Abstract: I will describe a framework for organizing patient-facing technologies to enable meaningful use from the patients’ perspective. The major premise is that success of healthcare reform is predicated on creating technology-enabled, proactive care teams coupled with informed and activated patients who are more engaged in their own care. Two technology-enabled research projects in primary care will be presented as exemplars of this model.

Power of Ecological Momentary Data with Modern Statistical Methods: New Questions, New Insights
Mariya Shiyko
Department of Counseling and Applied Educational Psychology
Northeastern University

Abstract: With technological innovations, electronic survey methods are becoming extensively used to study a broad spectrum of health-related behaviors (e.g., smoking, alcohol use, dieting). Repeated assessments of desired behaviors in real time as people live their lives presents a unique opportunity to understand etiology, dynamics, and contextual determinants that are ecologically sound and void of substantial biases. From the methodological perspective, resulting ecological momentary data (EM data) poses numerous analytical challenges, which are difficult to overcome with traditional statistical methods. This talk will focus on advances in statistical methodology for EM data and will concentrate on the nonparametric tools that allow exploration of complex patterns of relations measured with repeated observations. The talk is aimed at a broad non-technical audience and will introduce the statistical method in a context of empirical examples of smoking-cessation and quality of life research.

Quantifying Movement Disorder Symptoms by Voice
Max Little
Wellcome Trust/MIT Fellow

Abstract: For many progressive movement disorders such as Parkinson’s, it would be valuable in practice to detect the symptoms of the disease remotely, noninvasively, and objectively. Voice impairment is one of the primary symptoms of Parkinson’s. In this talk I’ll describe techniques that can be used to detect Parkinson’s and quantify the symptoms, using voice recordings alone. These algorithms achieve 99% overall accuracy in detecting the disease, and around 2% error in replicating the clinical symptom severity score on the Unified Parkinson’s Disease Rating Scale. I’ll also describe early results from the Parkinson’s Voice Initiative, a project that has captured a very large sample of voices from healthy controls and Parkinson’s subjects from around the world, using the standard telephone network.

Using Quantified Health Choices and Financial Rewards to Change Behavior in Adolescents and Young Adults
Dan Imler
Boston University/Boston Medical Center and Willpath

Abstract: Willpath helps people build healthier lifestyles and better bodies by quantifying real-world data about their small, daily health choices to give them encouraging feedback, financial reward and social validation when those choices are in line with evidence-based guidelines. Willpath was developed from clinical informatics research at Boston University which used an algorithm to determine the value of different health choices in relation to standard medical guidelines for a healthy lifestyle. This value metric is used to build a simple pathway to trusted outcomes that were previously too confusing, expensive, or just not fun.

Some of the things work for some of the people some of the time … Technology Enabled Behavior Change for Health
Doug McClure
Hea!thrageous

Healthrageous is a health service powered by a technology platform that helps individuals to shed unhealthy habits and create lasting health behavior change. The service and technology platform is based on early research conducted at the Partners HealthCare Center for Connected Health. That research showed the power of biometric feedback loops in health and wellness – something that has become popularly known as the quantified self. But, subsequent research and large scale market implementations have continued to show that behavior change comes from different motivations – inspirations, rewards, game play, communities, etc. By creating a synthesis of behavior change models that is powered by web, mobile and machine learning technologies; Healthrageous is creating a one-to–one, ultimately personalized experience that enables each of us to achieve our health goals.

Developing and Testing a Text Message Delivered Intervention for Smoking Cessation
Beth Bock
Brown University

There is growing interest in behavioral interventions delivered through mobile and computer platforms. However, using new platforms for treatment delivery requires developing new approaches and adapting older methods of providing treatment that are concordant with the ways in which technology is used, the preferences and use patterns of the target audience, and the limitations of the technology itself. TXT-2-Quit is an 8-week smoking cessation program designed for younger adult smokers who use text messaging. The experience of conducting TXT-2-Quit included qualitative formative work and a pilot randomized clinical trial. This presentation will highlight how methods of participant recruitment, our approach to providing counseling, and content and timing of the intervention were changed driven by the mobile platform.

The “App Store” for Health
Ken Mandl
Director, Intelligent Health Laboratory
Boston Children’s Hospital Informatics Program Associate Professor, Harvard Medical School

Abstract: We use electronic health records at work, but iPhones at home. Which do we prefer? Mandl presents a new approach to health information technology whereby healthcare gets an “App Store” of its own.

Despite the $48 billion dollar Federal investment in health information technology, doctors and hospitals have extraordinary difficulty implementing and using electronic health records (EHRs). How should Medicine construct an intelligent health system that facilitates rapid learning, innovation and transformation in wellness, health care, public health, and research?

Mandl presents the SMART Platform (www.smartplatforms.org), which capacitates EHRs and consumer-facing systems to run substitutable apps. As Mandl described in the New England Journal of Medicine, this software architecture supports creation and sustainability of an extensible ecosystem of apps, and stimulates a market for competition on value and price. The design principle of substitutability produces systems where a physician or researcher or a patient who wishes to add new functionality can simply delete an existing app and download a better one.

Mandl discusses the rapidly approaching tipping point in Medicine’s strategy for managing data and information for health and discovery.

The Communication Revolution and Health Inequalities in the 21st Century: Promises and Pitfalls
K. “Vish” Viswanath
Harvard School of Public Health

Abstract: The radical and transformative developments in Information and Communication Technologies (ICTs) offer unprecedented opportunities to enhance population and individual health. At the same time, given the current context in which these technologies are being deployed and used — inequities that characterize health AND communications across different social groups -race and ethnicity, social class, and nations raises important questions for public health research, policy and practice. This presentation will characterize communication inequalities, link them to health inequalities and offer exemplar solutions to address the inequalities.

PHI Open House
Short presentations from PHI Faculty
Details and video

Models of Behavior Change: Implications and Challenges
Wayne Velicer
University of Rhode Island

Abstract: Models of Behavior Change are different from Models of Behavior in four fundamental ways. First, models of behavior change focus on dynamic variables rather than static variable. Dynamic variables like psychosocial and behavioral variables are open to change. Static variables like past history and demographic variables cannot be easily modified. Second, the focus is on the individual evolutionary process. Models of Behavior Change need to model change over time. The nature of the change process and the pattern of change over time are critical. Third, models of behavior change need to be generalizable across behaviors. Models specific to a single behavior do little to advance science. Variables (manifest variables) will change across behaviors but constructs (latent variables) will not. Fourth, constructs that are curvilinear over time should be common in behavior change models. Typically, a person makes no effort to change a behavior, and then makes increasing effort, and then decreases effort as change becomes stable. Linear models, applied to curvilinear models will greatly underestimate the relationship. Models of behavior change will also result in a shift in focus for model testing with more emphasis on: (a) effect sizes estimation rather than traditional Null Hypothesis Testing; (b) Using methods like mediation analysis to identifying the mechanisms of behavior change; (c) Focusing on multivariate models to capture all aspects of the change process; (d) Focusing on longitudinal methods to include the temporal dimension; and (e) Testing theories through intervention development.

Thanksgiving Holiday – No Seminar

Wearable Systems to Monitor Responses to Rehabilitation Interventions
Paolo Bonato
Wyss Institute for Biologically Inspired Engineering, Harvard University
Department of Physical Medicine and Rehabilitation, Harvard Medical School

Abstract: The concept of monitoring individuals in the home and community settings was first proposed more than 50 years ago, when Holter monitoring was proposed (in the late 1940s) and later adopted (in the 1960s) as a clinical tool. However, technologies to fully enable such vision were lacking and only sporadic and rather obtrusive monitoring techniques were available for several decades. Over the past decade, we have witnessed a great deal of progress in the field of wearable sensors and systems. Advances in this field have finally provided the tools to implement and deploy technology with the capabilities required to achieve effective patients’ home monitoring. These technologies provide the tools to achieve early diagnosis of diseases such as congestive heart failure, prevention of chronic conditions such as diabetes, improved clinical management of neurodegenerative conditions such as Parkinson’s disease, and the ability to promptly respond to emergency situations such as seizures in patients with epilepsy and cardiac arrest in subjects undergoing cardiovascular monitoring. Current research efforts are focused on the development of systems enabling clinical applications. The current focus on developing and deploying wearable systems targeting specific clinical applications has the potential of leading to clinical adoption within the next five to ten years. In this talk, we will specifically focus on recent developments that appear to make the adoption of wearable technology attractive in the field of physical medicine and rehabilitation.

A New Wave of Patient-Centered Care: Technology-Supported Apprenticeship in the Management of Chronic Disease
John Moore
Massachusetts Institute of Technology

Abstract: Chronic disease is the most important cause of morbidity and mortality in developed countries, and the cost of its management under the current standard of care is unsustainable. The thesis of this talk is that apprenticeship as a paradigm of healthcare delivery, when supported by technological scaffolding and advanced communication tools, can improve the experience, clinical outcomes, and cost-effectiveness of chronic disease management. A software platform to support apprenticeship in the management of chronic disease, called CollaboRhythm, has been developed over the past 5 years. This platform is dramatically different than typical health information technology tools because it was designed to embrace the contribution of the patient and to support learning rather than to serve as a documentation and billing tool for physicians. Informed by research in computer supported cooperative work, it aims to maximally support effective collaboration between clinician and patient while being minimally intrusive in their interactions. Informed by pedagogical research, it aims to support techniques that have been validated in apprenticeship-based learning models such as scaffolding, modeling, articulation, coaching, reflection, and exploration. Randomized-controlled trials of apprenticeship applied to diabetes and hypertension management are being conducted to validate this new paradigm of healthcare delivery as well as the supporting CollaboRhythm platform.