NIH VideoCast - New Challenges for Big Data: Monitoring Behaviors in the Home and Environment

Skip Navigation

New Challenges for Big Data: Monitoring Behaviors in the Home and Environment

Loading... Loading video...

380 Views

Air date:	Thursday, January 24, 2013, 2:00:00 PM Time displayed is Eastern Time, Washington DC Local
Views:	Total views: 380 (45 Live, 335 On-demand)
Category:	BSSR - Behavioral and Social Sciences
Runtime:	01:13:13
Closed captioning:	Check box to display CC outside of the video Note: You can drag the captioning window around and resize it
Description:	BSSR Lecture There is an increasing focus on changing healthcare from being reactive and clinic- or hospital-based to being proactive and continuous, with an emphasis on interventions that make use of home monitoring and information/communications technology to facilitate scalable approaches for delivering care to the home. New developments in sensors, mobile apps and wireless devices have provided us with opportunities to track health behaviors. The new types of data streams that arise from continuous monitoring in the home environment include sleep quality metrics, activities of daily living, socialization measures, physical activity, gait measures and walking speed. Additionally, we also collect physiological home monitoring data used in disease management (blood glucose, peak flows, blood pressure, etc. ). These new types of data streams present many new challenges to “Big Data” modelers. The issues go beyond just thinking about volume of data and how to summarize or store it. For the behavioral monitoring in the home and environment there are now additional issues of 1) how to model context, bias and noise from signals derived from opportunistic low-cost sensors; 2) how to infer activities and behaviors from multiple sources, often with differing sampling rates and accuracies; and 3) how to manage privacy and security of sensitive data . Yet the opportunities for the discovery of new behavioral markers that will be useful in the management of health interventions are immense. With the continuous monitoring data we will be able to detect trends, using patients as their own controls, thus offering more sensitive and diagnostic measures by understanding what is normal for that individual. In addition, continuous or frequent data provides measures of variability in a signal, which is often diagnostic in itself. Finally, these new types of measures allow us to provide tailored health interventions with just-in-time feedback and support. These new monitoring techniques offer great promise for both reducing the cost of care and improving quality. However, there is an urgent need for research in data-analytic and data mining techniques to discover new clinical predictors, as well as research in how best to use individually tailored models of behavior to improve home health interventions. Holly B. Jimison, PhD is an IPA Health Scientist in NIH’s Office of Behavioral and Social Science Research, on loan from Oregon Health & Science University where she serves as an Associate Professor in the Department of Medical Informatics & Clinical Epidemiology. She received her Doctorate in Medical Information Sciences at Stanford University, with dissertation work on using computer decision models to tailor patient education materials to individuals. Dr. Jimison has both academic and industry experience in the design and evaluation of medical technologies. Her research is focused on consumer health informatics, with an emphasis on in-home monitoring and technology for successful aging. Her current projects include Big Data analytics for health behavior monitoring, technology for cognitive health coaching, and platforms to facilitate tailored home health interventions. Dr. Jimison is a Fellow of the American College of Medical Informatics, Past President of the Oregon Chapter of Health Information Management Systems Society, and serves on the Executive Council for Oregon’s Roybal Center for Aging & Technology.

NLM Title:	New challenges for big data : monitoring behaviors in the home and environment / Holly B. Jimison.
Author:	Jimison, Holly B. National Institutes of Health (U.S.)
Publisher:
Abstract:	(CIT): BSSR Lecture There is an increasing focus on changing healthcare from being reactive and clinic- or hospital-based to being proactive and continuous, with an emphasis on interventions that make use of home monitoring and information/communications technology to facilitate scalable approaches for delivering care to the home. New developments in sensors, mobile apps and wireless devices have provided us with opportunities to track health behaviors. The new types of data streams that arise from continuous monitoring in the home environment include sleep quality metrics, activities of daily living, socialization measures, physical activity, gait measures and walking speed. Additionally, we also collect physiological home monitoring data used in disease management (blood glucose, peak flows, blood pressure, etc. ). These new types of data streams present many new challenges to "Big Data" modelers. The issues go beyond just thinking about volume of data and how to summarize or store it. For the behavioral monitoring in the home and environment there are now additional issues of 1) how to model context, bias and noise from signals derived from opportunistic low-cost sensors; 2) how to infer activities and behaviors from multiple sources, often with differing sampling rates and accuracies; and 3) how to manage privacy and security of sensitive data . Yet the opportunities for the discovery of new behavioral markers that will be useful in the management of health interventions are immense. With the continuous monitoring data we will be able to detect trends, using patients as their own controls, thus offering more sensitive and diagnostic measures by understanding what is normal for that individual. In addition, continuous or frequent data provides measures of variability in a signal, which is often diagnostic in itself. Finally, these new types of measures allow us to provide tailored health interventions with just-in-time feedback and support. These new monitoring techniques offer great promise for both reducing the cost of care and improving quality. However, there is an urgent need for research in data-analytic and data mining techniques to discover new clinical predictors, as well as research in how best to use individually tailored models of behavior to improve home health interventions. Holly B. Jimison, PhD is an IPA Health Scientist in NIH"s Office of Behavioral and Social Science Research, on loan from Oregon Health & Science University where she serves as an Associate Professor in the Department of Medical Informatics & Clinical Epidemiology. She received her Doctorate in Medical Information Sciences at Stanford University, with dissertation work on using computer decision models to tailor patient education materials to individuals. Dr. Jimison has both academic and industry experience in the design and evaluation of medical technologies. Her research is focused on consumer health informatics, with an emphasis on in-home monitoring and technology for successful aging. Her current projects include Big Data analytics for health behavior monitoring, technology for cognitive health coaching, and platforms to facilitate tailored home health interventions. Dr. Jimison is a Fellow of the American College of Medical Informatics, Past President of the Oregon Chapter of Health Information Management Systems Society, and serves on the Executive Council for Oregon"s Roybal Center for Aging & Technology.
Subjects:	Data Mining Monitoring, Ambulatory Telemedicine
Publication Types:	Lecture Webcast

NLM Classification:	WB 142
NLM ID:	101601864
CIT Live ID:	12365
Permanent link:	https://videocast.nih.gov/watch=12365

Related Videos...