Using Serious Games to Measure Upper-Limb Myoelectric Pattern Recognition Prosthesis Control Performance in an At-Home Environment

Abstract

ABSTRACT Introduction Upper-limb prostheses have undergone substantial technological improvements in the last two decades, but user complaints and abandonment rates have hardly changed. This may follow from the use of laboratory-based tests when assessing new technology instead of measuring prosthesis use in at-home situations. Serious games might be used to assess prosthesis use at home. Objectives The aim of this study was to assess the feasibility, validity, and reliability of using serious games to measure myocontrol in at-home situations. Study Design The design of this study is a cohort study with repeated measurements. Methods All 10 participants (five males; seven transradial and three transhumeral defects) used pattern recognition (PR) to control their prosthesis. Two serious games were included in the PR software: Simon Says and In-The-Zone. Participants were instructed to use their prosthesis as they normally would for 2 weeks with the additional instruction to play at least 18 serious games (±90 minutes in total). Outcome measures for aspects of prosthesis use were wear time, months of prosthesis use, and user experiences, and serious games outcomes were success rate and completion time. Results For feasibility, all but one participant completed the minimum of 18 games, and user experiences were mostly positive. For validity, no relationships were found between the aspects of prosthesis use and performance in the serious games. For reliability, test-retest reliability for success rate was high (intraclass correlation [ICC], 0.79), but low for completion time (ICC, 0.18) in both games. Conclusions Feasibility and reliability of using serious games as an assessment instrument for prosthesis control in at-home situations appeared to be good; however, validity could not be established. The serious games should be improved to increase validity. Furthermore, more outcome measures that reveal aspects of prosthesis use should be considered. Clinical Relevance Statement Assessing myocontrol in at-home situations by using serious games is feasible and reliable, but validity needs further attention.