Current Low-Cost Video-Based Motion Analysis Options for Clinical Rehabilitation: A Systematic Review.

BACKGROUND: Physical therapists, as clinical human movement experts, must qualitatively evaluate patients' functional and biomechanical impairments. There are now low-cost 2- and 3-dimensional video measurement systems that can be used to increase the precision and reliability of these qualitative clinical assessments. PURPOSE: The purpose of this study was to systematically review current low-cost video-based methods for motion analysis compared with gold-standard 3-dimensional biomechanical methods. DATA SOURCES: Electronic searches were conducted until January 2018 within the following databases: MEDLINE via PubMed, Cumulative Index to Nursing and Allied Health Literature (CINAHL), Cochrane, Scopus, and the Institute of Electrical and Electronics Engineers. STUDY SELECTION: Studies designed to evaluate criterion-referenced validity and/or reliability of video-based motion analysis technologies within the last 20 years were included. English-language articles dealing with human rehabilitation were considered. DATA EXTRACTION: Data extraction was independently completed by 3 reviewers, and methodological quality was assessed using the 2018 Consensus-Based Standards for the Selection of Health Measurement Instruments checklist. Articles were organized for analysis on the basis of type of motion analyzed and category of each low-cost technology used. DATA SYNTHESIS: With 20 articles meeting selection criteria, 10 low-cost motion analysis platforms were presented, each examining different functional movement-dependent variables. Overall article quality was "low" or "very low" on the basis of Consensus-Based Standards for the Selection of Health Measurement Instruments scoring. Correlations between low-cost and 3-dimensional gold standard systems ranged widely from "poor" agreement (r = 0.025) to "strong" agreement (r = 0.992). Spatiotemporal gait parameters consistently outperformed planar joint angle data. Reliability was better measured than concurrent validity. A summary table was developed to assist clinicians in choosing which motions could potentially be measured accurately by each low-cost platform on the basis of current findings. LIMITATIONS: Databases available to researchers were more clinical/medical in nature, and this review was written from that clinically based perspective. Lack of standardized protocols and methodology within included studies was common, making generalizability difficult. CONCLUSIONS: Research attempting to validate newer low-cost movement analysis systems is limited in quality. Measurement of only certain variables should be considered when these tools are used. Further research is warranted, because these devices still have potential clinical utility for supplementing qualitative movement assessment with objective outcome measures.