Objective: To review the application of multivariate time series (MTS) analysis in assessing balance control and its potential for improving rehabilitation strategies.

Background: Balance control is a complex neuromuscular process involving multisystem coordination. Traditional assessments rely on static measures, limiting their ability to capture dynamic postural changes. Multivariate time series (MTS) analysis has emerged as a powerful computational approach for evaluating temporal dependencies in balance control, offering deeper insights into neuromuscular adaptation and stability.

Method: A systematic review of peer-reviewed studies was conducted across PubMed, IEEE Xplore, Scopus, and Web of Science databases up to March 2024. Inclusion criteria comprised studies employing MTS analysis in balance assessment across healthy and clinical populations. Extracted data included analytical techniques (e.g., Granger causality, autoregressive models, recurrent neural networks), sensor modalities, and clinical outcomes. A meta-analysis of predictive accuracy and diagnostic value was performed where applicable.

Results: A total of 28 studies (n = 1,040 participants) met inclusion criteria. Time-delay neural networks and autoregressive integrated models demonstrated high predictive accuracy (>85%) in identifying balance impairments. Studies utilizing wearable inertial sensors combined with MTS approaches achieved a 27% improvement in fall risk prediction compared to traditional methods. Dynamic cross-correlation analysis revealed significant neuromuscular lag patterns in patients with Parkinson’s disease and post-stroke conditions, supporting MTS as a valuable tool for early detection of balance deficits.

Conclusion: Multivariate time series analysis offers a transformative approach for assessing balance control, providing real-time, data-driven insights into postural stability and neuromuscular coordination. Its integration with wearable sensor technologies and machine learning models enhances predictive accuracy and clinical applicability. Future research should focus on developing standardized frameworks for MTS-based balance assessments and validating their effectiveness in diverse patient populations.

« Back to 2025 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/integration-of-multivariate-time-series-analysis-in-assessing-balance-control-a-comprehensive-review-of-current-research/