Georgia Tech Takes Leading Role in NIH-Funded Center for Mobile Sensor Data-to-Knowledge

Primary tabs

As part of a four-year, multi-university grant from the National Institutes of Health to develop innovative ways to use mobile and wearable sensors for health purposes, Georgia Tech will take a lead in developing data research, strategy, and commercial opportunities for the project.

The National Institutes of Health (NIH) announced on Oct. 9 that it has awarded $10.8 million in grants over the next four years to create the Mobile Sensor Data-to-Knowledge Center (MD2K). The center is part of the NIH Big Data to Knowledge (BD2K) initiative designed to support advances in research, policy, and training needed to use big data in biomedical research. To date, NIH has funded 12 “Centers of Excellence” to develop and deploy cutting-edge methods, tools, and other resources with big data.

The MD2K center will be based at University of Memphis, under the directorship of Santosh Kumar. Georgia Tech, which will receive about $1.2 million in grants for its role with the center, will serve as the lead of the center’s Data Science Research (DSR) core.

“The responsibility of the DSR team is to ensure that the center is successful from a research perspective and achieves its goals,” said Jim Rehg, professor in the School of Interactive Computing (IC). Rehg will serve as deputy director of the center and the lead scientist for data science research.

Gregory Abowd, a Regents and Distinguished Professor in IC, and Polo Chau, an assistant professor in the School of Computational Science and Engineering (CSE) and associate director of Tech’s Masters in Analytics program, also will conduct research funded through this grant.

For MD2K, the center assembles leading scientists from Cornell Tech, Georgia Tech, Northwestern, Ohio State, Rice, UCLA, UC San Diego, UC San Francisco, the University of Massachusetts Amherst, the University of Memphis, the University of Michigan, and Open mHealth, a non-profit organization.  Together, they aim to accelerate progress of predictive, preventive, personalized, participatory, and precision (P5) medicine by producing and making widely available open source, extensible, and standards-compliant big data analytics software for extracting information and actionable knowledge from mobile sensor data.

“Mobile sensors offer tremendous opportunities for accelerating biomedical discovery and optimizing care delivery,” said Kumar. “By resolving significant technological and scientific challenges related to the complexities of mobile sensor data, our team aims to lay the scientific foundations for realizing the vision of P5 Medicine with mobile sensors, and usher in the next generation of healthcare.”

The team will directly target two complex health conditions with high mortality risk – reducing hospital readmission in congestive heart failure (CHF) patients and preventing relapse in abstinent smokers. The MD2K tools, software, and training materials produced by the Center will be made widely available to researchers and clinicians, and will have the potential to impact other complex diseases such as asthma, substance abuse and obesity.

Rehg said his research will investigate the use of wearable cameras to obtain measures of health-related behavior. For example, he and his students will measure how often people are exposed to tobacco advertising, smokers, and other known potential cues for relapse. For cognitive heart failure patients, they will measure such behaviors as eating and drinking to see if the subject might be in danger of exceeding certain health restrictions, such as salt intake.

Abowd will assist Kumar and the executive committee on the strategic direction of the center, such as linking to the general computer science research community and advising on commercialization opportunities. His research will explore the use of on-body sensing to support automated and semi-automated detection of relevant human activities, such as eating and drinking.

Chau’s work will involve "explainable machine learning and data mining,” which combines scalable computation, intuitive interactive techniques, and data visualization to help users best understand data from mobile and wearable sensors.

Visit for additional information.


  • Workflow Status:
  • Created By:
    Brittany Aiello
  • Created:
  • Modified By:
    Fletcher Moore
  • Modified:


    No categories were selected.

Target Audience

    No target audience selected.