Accession Number : ADA616872


Title :   Detecting Gait Asymmetry with Wearable Accelerometers


Descriptive Note : Project rept.


Corporate Author : MASSACHUSETTS INST OF TECH LEXINGTON LINCOLN LAB


Personal Author(s) : Williamson, J R ; Dumas, A ; Hess, A R ; Patel, T ; Telfer, B A ; Fischl, K ; Butler, M J


Full Text : https://apps.dtic.mil/dtic/tr/fulltext/u2/a616872.pdf


Report Date : 18 Mar 2015


Pagination or Media Count : 38


Abstract : Gait asymmetry can be a useful indicator of a variety of medical and pathological conditions, including musculoskeletal injury (MSI), neurological damage associated with stroke or head trauma, and a variety of age-related disorders. Body-worn accelerometers provide the ability for real-time monitoring and detection of changes in gait asymmetry, thereby providing continuous real-time information about medical conditions and enabling timely interventions. We propose practical and robust algorithms for detecting gait asymmetry using features extracted from accelerometers attached to each foot. By registering simultaneous acceleration differences between the two feet, these asymmetry features provide robustness to a variety of confounding factors, such as changes in walking speed and load carriage. Because the algorithms require only summary statistics obtained from each frame (i.e., contiguous block) of accelerometer data, they can potentially be implemented in real-time physiological status monitoring systems, which may operate under severe limitations in power, computation, and communication bandwidth. We evaluate the algorithms on a data collection consisting of 24 subjects with multiple levels of induced gait asymmetries in both indoor and outdoor natural walking conditions. Changes in magnitude and pattern asymmetry features are sensitive to the sign and magnitude of gait asymmetry and provide the ability to detect and track asymmetries during continuous monitoring. By creating individualized background models from short data collections of normal walking, the algorithms are able to reliably detect asymmetrical walking induced by small ankle weights during short duration walking trials. Moreover, the background models and the test data are derived from both indoor and outdoor walking trials, where the outdoor walking contained uphill and downhill grades.


Descriptors :   *ACCELEROMETERS , *ALGORITHMS , *FEET , *WALKING , ACCELERATION , CASE STUDIES , DATA ACQUISITION , JOINTS(ANATOMY) , MONITORING , MUSCULOSKELETAL SYSTEM , NERVOUS SYSTEM , PATTERNS , REAL TIME , STATISTICS , WOUNDS AND INJURIES


Subject Categories : Anatomy and Physiology
      Numerical Mathematics
      Test Facilities, Equipment and Methods
      Navigation and Guidance


Distribution Statement : APPROVED FOR PUBLIC RELEASE