IEEE Conferences

The Future of Digital Health: Bridging Behavioral Science and Engineering with Intensive Longitudinal Assessment

Wednesday September 8, 15:00 - 16:20 UTC

Panel Outline:

• Overview of Network – Sy-Miin Chow
• Interpreting and Integrating Sensor Data in Intensive Longitudinal Digital Health Studies - Stephen Intille
• Engagement and Multi-Level Dynamic Modeling of Digital Health Data - Donna Spruijt-Metz
• Modeling Intensive Longitudinal Digital Health Data using Digital Twins Approach - Misha Pavel
• Clinical Applications and Impact: Intensive Longitudinal Assessment of Mental Health – Justin Baker
• Q&A

Advances in health behavior theories and the efficacy of health behavior interventions are limited by difficulties in invoking sustained health behavior changes within person across time. The Intensive Longitudinal Health Behavior Network (ILHBN) is a cooperative agreement network funded jointly by seven participating units within the National Institutes of Health to collaboratively study factors that influence key health behaviors in the dynamic environment of individuals, and ways to leverage intensive longitudinal data (ILD) collection and analytic methods to introduce innovations into long-standing behavioral theories and theory-driven behavior change interventions. The seven studies utilize a rich array of intensive longitudinal designs, data collection technologies (e.g., smartphones, wearables, video diaries), data types (e.g., ecological momentary assessments, location, accelerometry, physiological data, videos, images, and phone usage data), and analytic tools to study health behavior changes. This panel addresses several of the challenges in collecting and utilizing ILD.

The Temporal Influence of Movement and Exercise (TIME) Study is collecting phone and smartwatch data from over 250 people, each for a year, to investigate predictors of adoption and maintenance of behaviors related to physical activity, sedentary behavior, and sleep. Some of the methods being used will be described, along with challenges that have been encountered when interpreting and integrating sensor data for intensive longitudinal behavior measurement. The Dynamic Models of Behavior Study is a Micro-Randomized trial to increase physical activity in overweight but otherwise healthy adults. The project is collecting FitBit and phone data from 60 people for a year. A major challenge for this project, as well as all of the projects in the Network, is keeping participants engaged. We will address why engagement is key, and how it can be measured or captured using paradata, and how to tag and share this data across multiple projects.

The Bipolar Longitudinal Study (BLS) leverages smartphone technologies and data from recorded interviews, to establish robust behavioral markers associated with mania, depression, and psychosis experienced by collecting 100 person-years of multimodal data from at-risk individuals followed for up to 5 years. Our final speaker will address the challenges, benefits, and impact of utilizing ILD in a clinical setting. Finally, we will describe efforts to establish more robust approaches for translating domain knowledge about processes into computational models that account for theorized dynamics, and highlight some ways in which the cross-disciplinary collations from these projects have helped advance the field of digital health.

---

Panelists

Panel Chair: Sy-Miin Chow, PhD is a Professor in the Department of Human Development and Family Studies at the Pennsylvania State University and the Principal Investigator of the Emotions and Dynamic Systems Lab. Dr. Chow’s research focuses on the development and adaptation of modeling and analysis tools that are suited to evaluating linear and nonlinear dynamical systems models, including longitudinal structural equation models and state-space modeling techniques. Her current work involves using Kalman filter approaches and dynamical systems models to represent the dynamics of emotion regulation. Her longer term aim is to develop a broader repertoire of data-driven tools tailored toward analyzing the kinds of longitudinal data typically available in the social and behavioral sciences.

Donna Spruijt-Metz, MFA, PhD is a Research Professor in both Psychology and Preventive Medicine at the University of Southern California. She has focused for most of her career on mobile technologies to understand health-related behaviors as well as to prevent and treat obesity and diabetes in minority youth and families. Her research includes new work to redevelop psychosocial theories of behavior change using temporally dense, contextually rich continuous data. She also founded and co-directs USC mHealth Collaboratory with Dr. William Swartout, a leader in Artificial Intelligence.

Stephen Intille, PhD is an associate professor in the Khoury College of Computer Sciences and Bouvé College of Health Sciences at Northeastern University. His research focuses on the development of novel healthcare technologies that incorporate ideas from ubiquitous computing, user-interface design, pattern recognition, behavioral science, and preventive medicine. Intille is interested in human-computer interface technologies that measure and motivate health-related behaviors. In specific, how algorithms that recognize everyday activity can drive the development of interactive technologies that support healthy aging and well-being. Among his other research interests, he analyzes mobile technologies that permit longitudinal measurement of health behaviors and areas of human activity.

Misha Pavel, PhD holds a joint faculty appointment in Northeastern University’s Khoury College of Computer Sciences and Bouvé College of Health Sciences. His background comprises electrical engineering, computer science, and experimental psychology, and his research is focused on multi-scale computational modeling of behaviors and their control, with applications ranging from elder care to augmentation of human performance. Pavel uses these model-based approaches to develop algorithms transforming unobtrusive monitoring from smart homes and mobile devices to useful and actionable knowledge for diagnosis and intervention. Under the auspices of the Northeastern-based Consortium on Technology for Proactive Care, Pavel and his colleagues target technological innovations to support the development of economically feasible, proactive, distributed, and individual-centered healthcare. In addition, Pavel is investigating approaches to inferring and augmenting human intelligence using computer games, EEG, and transcranial electrical stimulation.

Justin T. Baker, MD, PhD, is the scientific director of the McLean Institute for Technology in Psychiatry (ITP) and director of the Laboratory for Functional Neuroimaging and Bioinformatics at McLean Hospital. He is also an assistant professor of psychiatry at Harvard Medical School. Dr. Baker’s research uses both large-scale studies and deep, multilevel phenotyping approaches to understand the nature and underlying biology of mental illnesses. He is a clinical psychiatrist with expertise in schizophrenia and bipolar spectrum disorders and other disorders of emerging adulthood. In 2016, Dr. Baker co-founded the ITP, a first-of-its-kind research and development center to foster tool development and novel applications of consumer technology in psychiatric research and care delivery.