GeoMed: Big Data for Health

The vision for GeoMed is to enable health applications by providing the capabilities to reveal and understand people's patterns of life. 

This vision parallels our vision for security with the difference that the environment and it's data informs the physiological and mental status of a person instead of simply serving as a backdrop on top of which the person's activities can be comprehended. PoCM2 is the main project unde the GeoMed umbrella, and is geared at exploring mobility monitoring in the home.

Other application will expand in moving from home to city and from a fixed sensor in the home sensor to using body sensors that can be used to monitor vital signs and network of sensors. In doing so we will have to address a number of challenges including data collection, indexing, analytics, visualization and privacy issues. PoCM2 – for Point of Care Remote Mobility Monitoring – is a novel Kinect-based assessment and monitoring tool for people with Parkinson’s disease (PD).

The system has the potential to reduce the risk of falls for people with PD and other patient populations with increased risk of falls such as stroke survivors or individuals with spinal cord injuries. We have developed a functional proof-of-concept system that was tested in a clinical setting for quantification of mobility data as an approach to estimate the impact of medication on people with PD, and ultimately recommend adjustments.

The PoCM2 system was developed as part of a 1 year funded project at IMSC through the Southern California Clinical and Translational Science Institute (SC CTSI) pilot grant mechanism and with support from the Alfred E. Mann Institute (AMI). In addition to the development activities that resulted in the PoCM prototype system, we included 15 participants with PD in focus groups (one before testing and one after). The focus groups, an important part of our collaborative process, revealed critical information that will be essential to the further development of PoCM. The prototype system uses the Microsoft Kinect® to provide clinicians with the ability to remotely measure a person’s mobility in the home and qualitatively evaluate stability, smoothness and coordination of movements. As shown in the figure, this tool provides a novel paradigm for mobility evaluation and monitoring compared to current practice methods that rely on subjective assessments and require the user to go into the clinic for those assessments. Existing automated technologies used for evaluation of musculoskeletal disorders are able to extract performance metrics with wearable sensors (e.g. accelerometers, gyroscopes, magnetometers) that have been clinically validated, however, they are intrusive and require users to turn the sensors on and off which can be difficult for some users with impaired motor control. Studies have reported on the use of Kinect for assessment in the home with a focus on feasibility. In contrast PoCM focuses on the improvement and extension of the data analytics to enhance its effective remote monitoring capabilities. The PoCM2 system is able to generate a robust representative skeletal sequence from the raw Kinect data that exhibits the user’s most consistent motion patterns. It also uses real-time analytics (including PCA-based data reduction and trajectory analysis) that are sensitive to the freezing of gait (FOG).


  • Banaei-Kashani F, Medioni G, Nguyen K, et al. PoCM2: Monitoring Mobility Disorders At Home Using 3D Visual Sensors and Mobile Sensors. In: Wireless Health; 2013; Baltimore, MD; 2013. 
  • Wang R, Medioni G, Winstein C, Blanco C. Home Monitoring Musculo-Skeletal Disorders with a Single 3D Sensor. In: International Workshop on Human Activity Understanding from 3D Data in conjunction with IEEE Conference on Computer Vision and Pattern Recognition (CVPR); 2013; Portland, OR; 2013.

Related publications

  • B. Pan, U. Demiryurek, C. Gupta, and C. Shahabi, Forecasting Spatiotemporal Impact of Traffic Incidents on Road Networks, IEEE International Conference on Data Mining (ICDM), Dallas, Texas, December 2013
  • Pan B, Demiryurek U, Shahabi C. Utilizing Real-World Transportation Data for Accurate Traffic Prediction. In: IEEE 12th International Conference on Data Mining (ICDM); 2012 10-13 Dec. 2012; 2012. p. 595-604.
  • Shahabi C, Marsh T, Yang K, et al. Immersidata Analysis: Four Case Studies. IEEE Computer 2007:45-52.
  • Yang K, Marsh T, Minyoung M, Shahabi C. Continuous archival and analysis of user data in virtual and immersive game environments. In; Proceeding CARPE '05 Proceedings of the 2nd ACM workshop on Continuous archival and retrieval of personal experiences. 2005:13- 22.