Demo DexterNet Wearable Sensor System
Mission
The mission of the PHAST Project is the creation of smart devices that promote awareness and understanding of personal activity spaces and how they affect health. Through the creation of wearable devices that integrate data-logging global positioning system receivers with personal and environmental sensors, our aim is to create integrated devices that can provide a wealth of individual-level data that can address major public health problems, such as the relationship between physical exercise and obesity, social interaction and the spread of infectious diseases, assessment of small-scale variations in exposures to airborne pollution, and environmental injustices. These devices allow for the spatial and temporal mapping of individual activity spaces and the identification of health risks associated with these spaces. Such information has the potential to improve Public Health by informing and promoting alternative environmental policies, community perspectives, and individual behaviors.
Research
CalFit
Keep track of your physical activity like never before on your mobile phone!
CalFit is a multi-user mobile application that monitors physical activity and encourages exercise through social interaction and competition. CalFit showcases the latest research in energy expenditure and activity tracking algorithms developed through collaboration between UC Berkeley Engineering and the School of Public Health. The application integrates data from various mobile phone sensors to track how and where exercise occurs. CalFit aims to fulfill two goals: 1) to promote healthier and more active lifestyles, and 2) to provide data on social and physical environments that conducive of physical activity that can inform future health policies and planning. CalFit runs on most of today’s Android smartphones, including the T-mobile G1, Motorola Droid, and Google Nexus One.
Screenshots of CalFit D (our datalogging version of CalFit):
Video of CalFit (our workout trainer version of CalFit):
 http://bsn.citris.berkeley.edu/media/berkeleyfit.mpg
 Wearable Sensor System (DexterNet)
DexterNet is a wearable sensor system that offers flexibility in applying a variety of sensors to personal health applications. Based on a network of wireless sensor motes,a Nokia N810 Internet tablet, and the SPINE software framework, the system can collect and process data from a variety of sensors including accelerometers/gyroscopes for motion and activity, global positioning system for geographic location, biosensors for heart and respiration rate, and air pollution data from gas sensors.
Kuryloski, P, Giani, A, Giannantonio, R, Gilani, K, Gravina, R, Seppä, V, Seto, E, Shia, V, Wang, C, Yan, P, Yang, A, Hyttinen, J, Sastry, S, Wicker, S, Bajcsy, R (2008), DexterNet: An Open Platform for Heterogeneous Body Sensor Networks and Its Applications. Paper presented at IPSN ’09, San Francisco, CA. [link]
GPS monitoring of mobilty patterns
Global positioning system receivers are sufficiently inexpensive and ubiquitous that they may be used to monitor social behaviors such as individual mobility patterns and person-person interactions that may be associated with the transmission of infectious diseases. Through the incorporation of these data into mathematical models of disease transmission we may better understand the role that these social behaviors play in the persistence and spread of disease.
Seto, E.Y.W. and Carlton, E.J. (2008) New Data on Social Connections for Parasitic Disease Modelling, Poster presented at Epidemics1, Asilomar, Pacific Grove, CA, December 2, 2008. [link to poster]
Gurarie, D. and Seto, E.Y.W. (2008) Connectivity Sustains Disease Transmission in Environments with Low Potential for Endemicity: Modeling Schistosomiasis with Hydrologic and Social Connectivities, Journal of the Royal Society Interface, doi:10.1098/rsif.2008.0265.
Seto, E.Y.W., Knapp, F., Zhong, B., Yang, C. (2007) The use of the global positioning system to assess individual-level water contact and time-activity patterns associated with schistosomiasis transmission, Geospatial Health, 1(2): 233-241. [link]
Sensor System for Monitoring Congestive Heart Failure Patients
In a collaborative project, the Berkeley DexterNet wearable sensor system was integrated with Vanderbilt's Trustworthy Health Information System (THIS) for a congestive heart failure patient monitoring application. Security was at the core of this integration, ensuring that data collected by wireless sensors are safely transmitted from mote to mobile basestation to the Intenet server application. The system currently supports data collected on activity level, blood pressure, and weight. The technology makes use of the Signal Processing In Note Environment (SPINE) framework, tinyos motes, secured communications, web interface, doctor-patient feedback, 3-axis accelerometer based energy expenditure calculation.
See our demonstration video:
Smart Health for Older AdultsFunded by NSF Smart Health and Wellbeing Program, researchers from UC Berkeley and Oregon Health Sciences University have embarked on a 4-year study to explore the use of home and personal sensor technologies to improve the health and wellness of elderly populations. Based on cutting-edge, yet practical technologies, we are working towards developing human-centric sensor systems, applications, and models that can assist the assist independent living of our older populations. The study's P.I. is Ruzena Bajcsy at UC Berkeley and Holly Jimison at OSHU. For more information and to learn about technology research for the elderly, see http://www.orcatech.org/
Validation of the CalFit system in environmental epidemiologic studiesFunded by NIH National Institutes of Environmental Health Sciences (NIEHS), Dr. Seto is collaborating with researchers in southern California (Pentz, USC) and in Spain (Nieuwenhuijsen, CREAL) to validate the smartphone-based physical activity and time-location monitoring of CalFit for ongoing cohort studies of the built environment and travel behavior. This 2-year study will also evaluate the use of CalFit to provide refined air pollution exposure estimates based on personal monitoring technologies. Dr. Seto serves as one of the two P.I.s of this project.
Collaborators
- UC Berkeley School of Public Health (Seto, Jerrett, Spear, Smith)
- UC Berkeley Electrical Engineering and Computer Science (Bajcsy)
- Children's Hospital Oakland Research Institute (Tester)
- Telecom Italia Wireless Sensor Lab (Sgroi)
- Vanderbilt
- Oregon Health Sciences University (Jimison)
Funding
- Center for Information Technology in the Interests of Society (CITRIS)
- NIH NIEHS
- NSF Smart Health and Wellbeing
- European Union
|
