Wearable smart glasses for first-person video analysis to evaluate nursing skills (Preprint)

BACKGROUND Comprehensive evaluations of skills and performance are necessary to improve the quality of care in nursing education. Various factors pose challenges to accurate evaluations, including high student–teacher ratio and observer bias. OBJECTIVE The present study is aimed at establishing a smart-glasses-based evaluation system as well as validation of its usability and feasibility. METHODS We used two cameras (smart glasses for first-person perspective and a web cam) to perform video analysis for skill and performance evaluations. Usability tests were conducted using quantitative and qualitative questionnaires for the usefulness and acceptability of the current evaluation system. Data from standardized checklists for three core nursing skills (IM, suction, and L-tube feeding) were used to analyze the inter-rater reliability, in addition to the Fleiss’ Kappa coefficient analysis. RESULTS A total of 22 participants (18 students and 4 reviewers) were recruited and completed the study. Apart from the physical discomfort component, the proposed smart-glasses-based evaluation system achieved good usability scores from both students and reviewers. The Fleiss’ Kappa values for the individual items ranged from 0.450 to 0.808, with the overall average value being 0.592. CONCLUSIONS The findings of this study verify the potential of first-person video evaluations using smart glasses to improve the accuracy and convenience of the skill evaluation process. However, issues regarding upgrading the device and need for guidelines for the roles of raters and students still remain.