College of Computer Science, Sichuan University, Chengdu, Sichuan 610000, China.
Department of Biomedical Informatics, Harvard Medical School, Boston, MA 02114, United States.
J Am Med Inform Assoc. 2024 Nov 1;31(11):2571-2582. doi: 10.1093/jamia/ocae224.
Conventional physical activity (PA) metrics derived from wearable sensors may not capture the cumulative, transitions from sedentary to active, and multidimensional patterns of PA, limiting the ability to predict physical function impairment (PFI) in older adults. This study aims to identify unique temporal patterns and develop novel digital biomarkers from wrist accelerometer data for predicting PFI and its subtypes using explainable artificial intelligence techniques.
Wrist accelerometer streaming data from 747 participants in the National Health and Aging Trends Study (NHATS) were used to calculate 231 PA features through time-series analysis techniques-Tsfresh. Predictive models for PFI and its subtypes (walking, balance, and extremity strength) were developed using 6 machine learning (ML) algorithms with hyperparameter optimization. The SHapley Additive exPlanations method was employed to interpret the ML models and rank the importance of input features.
Temporal analysis revealed peak PA differences between PFI and healthy controls from 9:00 to 11:00 am. The best-performing model (Gradient boosting Tree) achieved an area under the curve score of 85.93%, accuracy of 81.52%, sensitivity of 77.03%, and specificity of 87.50% when combining wrist accelerometer streaming data (WAPAS) features with demographic data.
The novel digital biomarkers, including change quantiles, Fourier transform (FFT) coefficients, and Aggregated (AGG) Linear Trend, outperformed traditional PA metrics in predicting PFI. These findings highlight the importance of capturing the multidimensional nature of PA patterns for PFI.
This study investigates the potential of wrist accelerometer digital biomarkers in predicting PFI and its subtypes in older adults. Integrated PFI monitoring systems with digital biomarkers would improve the current state of remote PFI surveillance.
传统的基于可穿戴传感器的身体活动 (PA) 指标可能无法捕捉 PA 的累积、从久坐到活跃的转变以及多维模式，从而限制了预测老年人身体功能障碍 (PFI) 的能力。本研究旨在使用可解释的人工智能技术，从腕部加速度计数据中识别出独特的时间模式并开发新的数字生物标志物，用于预测 PFI 及其亚型。
使用来自国家健康老龄化趋势研究 (NHATS) 的 747 名参与者的腕部加速度计流式数据，通过时间序列分析技术-Tsfresh 计算 231 个 PA 特征。使用 6 种机器学习 (ML) 算法和超参数优化，为 PFI 及其亚型（行走、平衡和四肢力量）开发预测模型。使用 Shapley Additive exPlanations 方法解释 ML 模型并对输入特征的重要性进行排序。
时间分析显示，PFI 与健康对照组之间的 PA 峰值差异出现在上午 9:00 至 11:00 之间。表现最佳的模型（梯度提升树）在结合腕部加速度计流式数据 (WAPAS) 特征和人口统计学数据时，曲线下面积评分为 85.93%，准确率为 81.52%，灵敏度为 77.03%，特异性为 87.50%。
包括变化分位数、傅里叶变换 (FFT) 系数和聚合 (AGG) 线性趋势在内的新型数字生物标志物在预测 PFI 方面优于传统 PA 指标。这些发现强调了捕捉 PA 模式多维性质对于预测 PFI 的重要性。
本研究调查了腕部加速度计数字生物标志物在预测老年人 PFI 及其亚型方面的潜力。具有数字生物标志物的综合 PFI 监测系统将改善当前的远程 PFI 监测状态。
