Background Knee osteoarthritis (OA) is a debilitating condition that compromises mobility and exacerbates knee pain, necessitating accurate and accessible diagnostic tools. Traditional motion capture technology, while effective, is often cost-prohibitive and limited to laboratory settings. In response, we developed a smartphone-based approach utilizing spatiotemporal analysis of joint angular velocities and angles in sit-to-stand (STS) motion to detect symptomatic knee OA. Method We analyzed 2063 sagittal-viewed sit-to-stand motion videos from 309 participants and proposed a STS-D Index based on a deep learning model, STS Dynamics Net, which provides a nuanced quantification of joint dynamics and temporal interactions in trunk, knee, and ankle angles and velocities for detection of symptomatic knee OA. Results Here we show that joint angular velocities are a robust spatiotemporal biomarker for symptomatic knee OA detection (AUC 0.7759 ± 0.0219), not only do they outperform the STS pace (AUC 0.6554 ± 0.0268, p = 8.6<InlineEquation ID="IEq1"> <EquationSource Format="TEX">$\times$</EquationSource> <EquationSource Format="MATHML"><math> <mo>×</mo> </math></EquationSource> </InlineEquation>10-5) and maximum trunk angle (AUC 0.7025 ± 0.0253, p = 2.1<InlineEquation ID="IEq2"> <EquationSource Format="TEX">$\times$</EquationSource> <EquationSource Format="MATHML"><math> <mo>×</mo> </math></EquationSource> </InlineEquation>10-3) in diagnostic accuracy and rival the performance of gold-standard 3D marker-based systems (AUC 0.7855 ± 0.0229), but they also show significant correlations with WOMAC sub-scores (p < 0.0001). Furthermore, our analysis reveals a significant correlation between angular velocities and muscle volumes and fat-to-muscle ratios in the quadriceps and hamstrings, underscoring the role of muscle weakness in knee OA pathogenesis. Conclusions This innovative approach has the potential to revolutionize knee OA detection, enabling reliable, cost-effective, and self-administered assessments in community settings and bridging the gap in accessible healthcare monitoring.

Smartphone-derived joint angular velocities in sit-to-stand motion provide a spatiotemporal marker for symptomatic knee osteoarthritis

期刊论文
出版模式：

开放获取
发表日期：

2026年3月24日

查看全文

Lok Chun Chan,
Jin Yan,
Yuli Charlie Zhang,
Tianshu Jiang,
Alex Yuning Zhang,
Ho Hin Toby Li,
Billy So,
Wei Huang,
Yongping Zheng,
Ping-Keung Chan,
Chunyi Wen

摘要

Background

Knee osteoarthritis (OA) is a debilitating condition that compromises mobility and exacerbates knee pain, necessitating accurate and accessible diagnostic tools. Traditional motion capture technology, while effective, is often cost-prohibitive and limited to laboratory settings. In response, we developed a smartphone-based approach utilizing spatiotemporal analysis of joint angular velocities and angles in sit-to-stand (STS) motion to detect symptomatic knee OA.

Method

We analyzed 2063 sagittal-viewed sit-to-stand motion videos from 309 participants and proposed a STS-D Index based on a deep learning model, STS Dynamics Net, which provides a nuanced quantification of joint dynamics and temporal interactions in trunk, knee, and ankle angles and velocities for detection of symptomatic knee OA.

Results

Here we show that joint angular velocities are a robust spatiotemporal biomarker for symptomatic knee OA detection (AUC 0.7759 ± 0.0219), not only do they outperform the STS pace (AUC 0.6554 ± 0.0268, p = 8.6 $\times$ $\times$ 10^-5) and maximum trunk angle (AUC 0.7025 ± 0.0253, p = 2.1 $\times$ $\times$ 10^-3) in diagnostic accuracy and rival the performance of gold-standard 3D marker-based systems (AUC 0.7855 ± 0.0229), but they also show significant correlations with WOMAC sub-scores (p < 0.0001). Furthermore, our analysis reveals a significant correlation between angular velocities and muscle volumes and fat-to-muscle ratios in the quadriceps and hamstrings, underscoring the role of muscle weakness in knee OA pathogenesis.

Conclusions

This innovative approach has the potential to revolutionize knee OA detection, enabling reliable, cost-effective, and self-administered assessments in community settings and bridging the gap in accessible healthcare monitoring.