Stacy Marsella, Professor of Computer and Information Science and Psychology
Marsella, a leader in the field of human-computer interaction, is opening whole new possibilities for computers to promote healthy behavior and help us learn more about how and why people act as they do.
Using a combination of psychological models and machine learning tools—including technologies pioneered by Marsella and his research partners—he is creating virtual humans that interact with people in ways that are more human than virtual.
Picking up on voice and visual cues—a hesitant reply or a downcast gaze, for example—Marsella’s virtual humans can infer the emotions of a human subject and respond with meaningful questions and comments, while using the kind of realistic facial expressions and hand gestures that are lubricants to human interaction.
Marsella and his team are making computational models of behavior easier to assess by putting them into human form. These virtual humans also have practical applications as coaches in tasks such as teaching medical students how to break bad news to patients and helping gay men avoid risky behaviors.
Andrea Parker, assistant professor of computer and information science and health sciences
Parker is shaping the field of personal health informatics—the use of technology to influence and inform the choices people make about their health.
She begins by examining disparities—such as limited access to nutritious food—that affect people with limited economic resources and elevate their risk for chronic health conditions. Then she applies her own interactive solutions to help people overcome those disparities.
For example, Parker is collaborating with colleagues at Northeastern to create digital programs—a mix of mobile applications, web-based games, and social media—to encourage low-income families to increase their weekly exercise and teenagers to become peer advocates for good nutrition.
To be effective, the technology must align with the communities she’s trying to reach. It has to “feel authentic,” she says, to connect naturally with people and their lives, in ways that energize and motivate.
Gary Young, professor of strategic management and healthcare systems
Young’s landmark study of U.S. hospitals’ provision of community benefits provides the hard data that will fuel debate on hospital spending priorities, just as provisions of the Affordable Care Act affecting those priorities begin to take effect.
The study, published in April in the New England Journal of Medicine, is based on newly available information about the services and activities that hospitals offer in their communities—benefits for which they receive no reimbursement.
It reveals wide disparities in hospital expenditures for those offerings and raises questions about how well hospitals are responding to ACA priorities for preventive care and population health.
Young and his interdisciplinary team at the Center for Health Policy and Healthcare Research found that some hospitals devote more than 20 percent of their operating budget to community benefits, while others contribute less than 1 percent. The study showed that, regardless of spending levels, the provided benefits were unlikely to include preventive medicine and wellness education.
Stephen Intille, associate professor of health sciences and computer and information science
A leader in the field of personal health informatics, Intille is taking a unique approach to promoting healthy behavior—adapting a device nearly everyone owns, the smartphone, to assist with the motivating.
Intille and his team are building on the smartphone’s capabilities to help us track data related to the choices we make affecting our health. They are creating tools to help us understand our choices—about eating, exercise, even socializing—and developing mathematical models that enable computers to synthesize and respond to the information almost immediately.
One goal: an app that serves as a personal health coach, capable of sending instantaneous messages that reinforce positive health behaviors—an encouraging voice message from a spouse, for example.
The results of Intille’s trials will shape fields like exercise science, sleep science, and weight loss and nutrition—issues affecting everyone, from teenagers fighting obesity to aging baby boomers.
Timothy Bickmore, associate professor of computer and information science
Bickmore is using technology to help patients manage their own healthcare in a way no one else has. Meet Tanya, an avatar or “relational agent” in Bickmore’s phrase, who can serve as a nurse and personal health advocate.
By studying the behavior of real nurses and then turning his observations into complex computational algorithms, Bickmore is able to create avatars that show empathy and converse naturally with patients. They can access a patient’s medical records and provide information about drug treatments.
Avatars have unlimited time to walk patients through often confusing postclinic procedures, such as when to take their medication and how to dress a wound—abilities relevant to the critical issue of hospital readmissions.
In fact, a majority of patients involved in early clinical trials—particularly those with limited health and computer literacy—reported feeling more at ease interacting with avatars like Tanya than with live nurses.
Rupal Patel, associate professor of computer and information science and speech language pathology and audiology
Healthy adults with fully developed vocal systems convey information by producing speech and changing the melody of their voice.
But children and adults with severe speech-motor disorders tend to rely more heavily on melodic cues, such as volume and duration. Patel uses her understanding of speech melody to create computational tools that can dramatically improve a disordered speaker’s ability to interact with the world.
In one project, Patel overlays meolodic fluctuations from a disordered speaker’s voice with a sentence spoken by a healthy donor of the same demographic. By merging the two signals, she creates a novel synthetic voice that conveys the user’s personal identity.
For children learning how to read, Patel also develops digital tools with visual cues—such as a rising and falling line—that signal pitch changes. Research suggests that by understanding the melody of speech earlier, children may achieve greater reading comprehension.
Carmen Sceppa, associate professor of health sciences
Sceppa is in the vanguard of health scientists who want to transform the way we think about exercise, from a personal choice to preventive medicine.
Chronic health conditions that afflict two-thirds of Americans have roots that go deeper than people’s individual medical and behavioral profiles, says Sceppa. Community factors—environment, economics, and culture—can inhibit regular physical activity.
Changing the terms of the conversation—proving that daily exercise is as effective in maintaining health as any drug—can help overcome those factors, and give the medical community a push in the direction of prescribing a workout instead of a pill.
Sceppa is collaborating with colleagues in nutrition and public health to develop the evidence-based models needed to achieve that transformation.
For example, she and her team successfully tested a community-based, largely peer-led strength-training program for older adults in Boston, and plan to build on it as a model for larger studies.