Accuracy and Reliability of Wearable Devices,Image Analysis Software,and Smartphone Apps for Hand Goniometry: A Systematic Review

Abstract

Several wearable devices, image analysis systems, and smartphone apps have been developed to increase the accuracy and reliability of hand range of motion measurement. While their performances have been assessed independently, there has been no comparison of devices across different groups. This study aims to compare the accuracy, reliability, and telehealth applicability of hand goniometry devices across methodological classes. PubMed, Embase, Scopus, and CINAHL were queried for studies evaluating the accuracy and reliability of hand goniometry technologies. The inquiry focused on tools that were directly compared to manual goniometry. 13 studies met inclusion criteria, representing 2 wearable devices, 7 image analysis systems, and 4 smartphone apps. Wearable devices had the highest accuracy (mean difference [MD] < 1^o) but had low telehealth integration scores (TIS = 38 out of 50). Image analysis systems had high reliability (interclass correlation coefficient [ICC] > .67) and telehealth usability (TIS = 42) but lacked accuracy (MD 1°–35°). Similarly, smartphone apps had low accuracy (MD 1°–12°) but high reliability (ICC > 0.83). No single device class outperforms others across all metrics. Future studies should focus on creating technologies that combine the strengths of multiple device classes.