Brand Yonatan E, Buchman Aron S, Kluge Felix, Palmerini Luca, Becker Clemens, Cereatti Andrea, Maetzler Walter, Vereijken Beatrix, Yarnall Alison J, Rochester Lynn, Del Din Silvia, Mueller Arne, Hausdorff Jeffrey M, Perlman Or
School of Biomedical Engineering, Tel Aviv University, Tel Aviv, Israel.
Center for the Study of Movement, Cognition and Mobility, Neurological Institute, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel.
medRxiv. 2025 May 21:2025.05.21.25328061. doi: 10.1101/2025.05.21.25328061.
Physical activity and mobility are critical for healthy aging and predict diverse health outcomes. While wrist-worn accelerometers are widely used to monitor physical activity, estimating gait metrics from wrist data remains challenging. We extend ElderNet, a self-supervised deep-learning model previously validated for walking-bout detection, to estimate gait metrics from wrist accelerometry. Validation involved 819 older adults (Rush-Memory- and-Aging-Project) and 85 individuals with gait impairments (Mobilise-D), from six medical centers. In Mobilise-D, ElderNet achieved an absolute error of 8.82 cm/s and an intra-class correlation of 0.87 for gait speed, outperforming state-of-the-art methods (p < 0.001) and models using a lower-back sensor. ElderNet outperformed (percentage error; p < 0.01) competing approaches in estimating cadence and stride length, and better (p < 0.01) classified mobility disability (AUC = 0.80) than conventional gait or physical activity metrics. These results render ElderNet a scalable tool for gait assessment using wrist-worn devices in aging and clinical populations.
身体活动和移动能力对健康老龄化至关重要，并能预测多种健康结果。虽然腕部佩戴的加速度计被广泛用于监测身体活动，但从腕部数据估计步态指标仍然具有挑战性。我们扩展了ElderNet，这是一种先前已通过步行检测验证的自监督深度学习模型，用于从腕部加速度测量中估计步态指标。验证涉及来自六个医疗中心的819名老年人（拉什记忆与衰老项目）和85名步态受损个体（Mobilise-D）。在Mobilise-D中，ElderNet在步态速度方面实现了8.82 cm/s的绝对误差和0.87的类内相关性，优于现有方法（p < 0.001）以及使用下背部传感器的模型。在估计步频和步幅方面，ElderNet优于竞争方法（百分比误差；p < 0.01），并且在对行动障碍进行分类方面（AUC = 0.80）比传统步态或身体活动指标表现更好（p < 0.01）。这些结果使ElderNet成为一种可扩展的工具，用于在老龄化和临床人群中使用腕部佩戴设备进行步态评估。
