4th International ICST Conference on Pervasive Computing Technologies for Healthcare

Research Article

Automated administration of the Wolf Motor Function Test for post-stroke assessment

Download517 downloads
  • @INPROCEEDINGS{10.4108/ICST.PERVASIVEHEALTH2010.8903,
        author={Eric Wade and Avinash Rao Parnandi and Maja J. Mataric},
        title={Automated administration of the Wolf Motor Function Test for post-stroke assessment},
        proceedings={4th International ICST Conference on Pervasive Computing Technologies for Healthcare},
        proceedings_a={PERVASIVEHEALTH},
        year={2010},
        month={6},
        keywords={Application software Automatic testing Cameras Computer science Instruments Medical treatment Neuroscience Standardization Timing Wearable sensors},
        doi={10.4108/ICST.PERVASIVEHEALTH2010.8903}
    }
    
  • Eric Wade
    Avinash Rao Parnandi
    Maja J. Mataric
    Year: 2010
    Automated administration of the Wolf Motor Function Test for post-stroke assessment
    PERVASIVEHEALTH
    ICST
    DOI: 10.4108/ICST.PERVASIVEHEALTH2010.8903
Eric Wade1,*, Avinash Rao Parnandi2,*, Maja J. Mataric3,*
  • 1: Department of Biokinesiology and Physical Therapy, University of Southern California, Los Angeles, CA, USA
  • 2: Department of Electrical Engineering, University of Southern California, Los Angeles, CA, USA
  • 3: Computer Science and Neuroscience, University of Southern California, Los Angeles, CA, USA
*Contact email: ericwade@usc.edu, parnandi@usc.edu, mataric@usc.edu

Abstract

The advent of new health sensing technologies has presented us with the opportunity to gain richer data from patients undergoing clinical interventions. Such technologies are particularly suited for applications requiring temporal accuracy. The Wolf Motor Function Test (WMFT) is one such application. This assessment is an instrument used to determine functional ability of the paretic and non-paretic limbs in individuals post-stroke. It consists of 17 tasks, 15 of which are scored according to both time and a functional ability scale. We propose a technique that uses wearable sensors and performance sensors to estimate the timing of seven of these tasks. We have developed a sensing framework and an algorithm to automatically detect total movement time. We have validated the system's accuracy on the seven selected WMFT tasks. We also suggest how this framework can be adapted to the remaining tasks.