肢体冲击波心动图与血压关联的数据挖掘研究。

Data mining investigation of the association between a limb ballistocardiogram and blood pressure.

机构信息

Department of Mechanical Engineering, University of Maryland, College Park, MD, United States of America.

出版信息

Physiol Meas. 2018 Aug 1;39(7):075009. doi: 10.1088/1361-6579/aacfe1.

DOI:10.1088/1361-6579/aacfe1

PMID:29952758

Abstract

OBJECTIVE

To investigate the association between a limb ballistocardiogram (BCG) and blood pressure (BP) based on data mining.

APPROACH

During four BP-perturbing interventions, the BCG and reference BP were measured from 23 young, healthy volunteers using a custom-manufactured wristband equipped with a MEMS accelerometer and a commercial continuous BP measurement device. Both timing and amplitude features in the wrist BCG waveform were extracted, and significant features predictive of diastolic (DP) and systolic (SP) BP were selected using stepwise linear regression analysis. The selected features were further compressed using principal component analysis to yield a small set of DP and SP predictors. The association between the predictors thus obtained and BP was investigated by multivariate linear regression analysis.

MAIN RESULTS

The predictors exhibited a meaningful association with BP. When three most significant predictors were used for DP and SP, a correlation coefficient of r = 0.75 ± 0.03 (DP) and r = 0.75 ± 0.03 (SP), a root-mean-squared error (RMSE) of 7.4 ± 0.6 mmHg (DP) and 10.3 ± 0.8 mmHg (SP), and a mean absolute error (MAE) of 6.0 ± 0.5 mmHg (DP) and 8.3 ± 0.7 mmHg (SP) were obtained across all interventions (mean ± SE). The association was consistent in all the individual interventions (r ⩾ 0.68, RMSE ⩽ 5.7 mmHg, and MAE ⩽ 4.5 mmHg for DP as well as r ⩾ 0.61, RMSE ⩽ 7.9 mmHg, and MAE ⩽ 6.4 mmHg for SP on the average). The minimum number of requisite predictors for robust yet practically realistic BP monitoring appeared to be three. The association between predictors and BP was maintained even under regularized calibration (r = 0.63 ± 0.05, RMSE = 9.3 ± 0.8 mmHg, and MAE = 7.6 ± 0.7 mmHg for DP as well as r = 0.60 ± 0.05, RMSE = 14.7 ± 1.4 mmHg, and MAE = 11.9 ± 1.1 mmHg for SP (mean ± SE)). The requisite predictors for DP and SP were distinct from each other.

SIGNIFICANCE

The results of this study may provide a viable basis for ultra-convenient BP monitoring based on a limb BCG alone.

摘要

目的

基于数据挖掘研究肢体心动描记图（BCG）与血压（BP）之间的关系。

方法

在 4 次 BP 干扰干预中，使用定制的腕带，该腕带配备了 MEMS 加速度计和商业连续 BP 测量设备，对 23 名年轻健康志愿者进行了 BCG 和参考 BP 的测量。从腕部 BCG 波形中提取了时间和幅度特征，并使用逐步线性回归分析选择了与舒张压（DP）和收缩压（SP）相关的显著特征。使用主成分分析进一步压缩所选特征，得到一组较小的 DP 和 SP 预测因子。使用多元线性回归分析研究了由此获得的预测因子与 BP 之间的关系。

主要结果

预测因子与 BP 具有有意义的关联。当使用三个最显著的预测因子进行 DP 和 SP 时，相关性系数 r=0.75±0.03（DP）和 r=0.75±0.03（SP），均方根误差（RMSE）为 7.4±0.6mmHg（DP）和 10.3±0.8mmHg（SP），平均绝对误差（MAE）为 6.0±0.5mmHg（DP）和 8.3±0.7mmHg（SP），在所有干预中都得到了验证（平均值±SE）。这种关联在所有单独的干预中都是一致的（r≥0.68，RMSE≤5.7mmHg，MAE≤4.5mmHg 用于 DP 以及 r≥0.61，RMSE≤7.9mmHg，MAE≤6.4mmHg 用于 SP）。似乎最少需要三个预测因子才能进行稳健但实际可行的 BP 监测。即使在正则化校准下，预测因子与 BP 之间的关联仍然保持不变（r=0.63±0.05，RMSE=9.3±0.8mmHg，MAE=7.6±0.7mmHg 用于 DP 以及 r=0.60±0.05，RMSE=14.7±1.4mmHg，MAE=11.9±1.1mmHg 用于 SP（平均值±SE））。DP 和 SP 的必需预测因子彼此不同。