Vítečková Slávka, Horáková Hana, Poláková Kamila, Krupička Radim, Růžička Evžen, Brožová Hana
Faculty of Biomedical Engineering, Czech Technical University in Prague, Kladno, Czech Republic.
Department of Neurology and Center of Clinical Neuroscience, 1st Faculty of Medicine and General University Hospital in Prague, Charles University, Prague, Czech Republic.
PeerJ. 2020 May 22;8:e8835. doi: 10.7717/peerj.8835. eCollection 2020.
Nowadays, the most widely used types of wearable sensors in gait analysis are inertial sensors. The aim of the study was to assess the agreement between two different systems for measuring gait parameters (inertial sensor vs. electronic walkway) on healthy control subjects (HC) and patients with Parkinson's disease (PD).
Forty healthy volunteers (26 men, 14 women, mean age 58.7 ± 7.7 years) participated in the study and 24 PD patients (19 men, five women, mean age 62.7 ± 9.8 years). Each participant walked across an electronic walkway, GAITRite, with embedded pressure sensors at their preferred walking speed. Concurrently a G-Walk sensor was attached with a semi-elastic belt to the L5 spinal segment of the subject. Walking speed, cadence, stride duration, stride length, stance, swing, single support and double support phase values were compared between both systems.
The Passing-Bablock regression slope line manifested the values closest to 1.00 for cadence and stride duration (0.99 ≤ 1.00) in both groups. The slope of other parameters varied between 0.26 (double support duration in PD) and 1.74 (duration of single support for HC). The mean square error confirmed the best fit of the regression line for speed, stride duration and stride length. The y-intercepts showed higher systematic error in PD than HC for speed, stance, swing, and single support phases.
The final results of this study indicate that the G-Walk system can be used for evaluating the gait characteristics of the healthy subjects as well as the PD patients. However, the duration of the gait cycle phases should be used with caution due to the presence of a systematic error.
如今，步态分析中使用最广泛的可穿戴传感器类型是惯性传感器。本研究的目的是评估两种不同系统在测量健康对照受试者（HC）和帕金森病（PD）患者步态参数（惯性传感器与电子步道）之间的一致性。
40名健康志愿者（26名男性，14名女性，平均年龄58.7±7.7岁）参与了本研究，还有24名PD患者（19名男性，5名女性，平均年龄62.7±9.8岁）。每位参与者以其偏好的步行速度走过一条装有嵌入式压力传感器的电子步道GAITRite。同时，使用一条半弹性腰带将G-Walk传感器附着在受试者的L5脊柱节段上。比较了两个系统之间的步行速度、步频、步幅持续时间、步幅长度、站立期、摆动期、单支撑期和双支撑期值。
两组中，Passing-Bablock回归斜率线显示步频和步幅持续时间的值最接近1.00（0.99≤1.00）。其他参数的斜率在0.26（PD患者的双支撑持续时间）和1.74（HC患者的单支撑持续时间）之间变化。均方误差证实了速度、步幅持续时间和步幅长度的回归线拟合最佳。在速度、站立期、摆动期和单支撑期方面，y轴截距显示PD患者的系统误差高于HC患者。
本研究的最终结果表明，G-Walk系统可用于评估健康受试者以及PD患者的步态特征。然而，由于存在系统误差，步态周期各阶段的持续时间应谨慎使用。
