Longitudinal wearable tremor measurement system with activity recognition algorithms for upper limb tremor. Conference Paper uri icon

abstract

  • Assessments of tremor characteristics by movement disorder physicians are usually done at single time points in clinic settings, so that the description of the tremor does not take into account the dependence of the tremor on specific behavioral situations. Moreover, treatment-induced changes in tremor or behavior cannot be quantitatively tracked for extended periods of time. We developed a wearable tremor measurement system with tremor and activity recognition algorithms for long-term upper limb behavior tracking, to characterize tremor characteristics and treatment effects in their daily lives. In this pilot study, we collected sensor data of arm movement from three healthy participants using a wrist device that included a 3-axis accelerometer and a 3-axis gyroscope, and classified tremor and activities within scenario tasks which resembled real life situations. Our results show that the system was able to classify the tremor and activities with 89.71% and 74.48% accuracies during the scenario tasks. From this results, we expect to expand our tremor and activity measurement in longer time period.

name of conference

  • 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)

published proceedings

  • Annu Int Conf IEEE Eng Med Biol Soc

author list (cited authors)

  • Jeonghee Kim, .., Parnell, C., Wichmann, T., & DeWeerth, S. P.

complete list of authors

  • Parnell, Claire||Wichmann, Thomas||DeWeerth, Stephen P

publication date

  • January 1, 2016 11:11 AM