관성센서로부터 추출되어진 파라미터 기반의 보행대칭 평가

Abstract

Although gait symmetry is being evaluated and reported in the literature with increasing frequency, there is still no generally accepted standard for assessing symmetry, making it difficult to compare studies and establish criteria to guide clinical decision-making. The objective of this study is to assess gait symmetry using spatio-temporal parameters including kinematic information (joint angle) based on gait phase by attaching seven Attitude and Heading Reference System (AHRS) modules. The seven in-house AHRS modules were designed to be attached to lower limbs (lumbar, left/right thigh, shank and foot), with each module sending Euler angle values calculated using a gradient descent algorithm to a PC via wireless transmission. In particular, the AHRS modules attached to each foot were designed to transmit 3-axis accelerometer signals as well as Euler angles in order to calculate spatial parameters. 8 spatio-temporal parameters (Gait cycle time, Stance, Swing, Pre-swing phase, Single, double support, velocity, distance) and 3 kinematic parameters (Hip, knee and ankle angle) are extracted from data that are transmitted from 7 AHRS modules and symmetry was measure with the Gait asymmetry index. In case of kinematic parameters, Gait cycle is divided into 6 gait phases on the basis of temporal parametric information, and the joint angle in 3-dimensional space are employed for analysis in sagittal, frontal and transverse plane. In this study, 15 health adults in their 20s and 30s participated in an experiment with three conditions for gait analysis: non-assisted walking (I); changing the height of the heel for one foot using a Rom-walker to induce abnormal gait (Rom-walker assisted walking (II)); and restricting ankle joint by adjusting ankle plantar-flexion additionally to condition II (Rom-walker assisted walking with 15-degree slope (III)). The results of analysis showed that, regarding temporal parameters (gait cycle time, stance, swing, pre-swing phase, single support time, and double support time), the Gait Asymmetry (GA) index was larger in Rom-walker assisted walking (II) than in non-assisted walking (I), although the difference was not significant. In addition, the GA index was significantly larger in Rom-walker assisted walking with 15-degree slope (III), where the ankle joint was restricted, than in non-assisted walking (I) (P<0.01). Different conditions of Rom-walker assisted walking also showed a significant difference (P<0.01). Regarding double support time, both Rom-walker assisted walking (II) and Rom-walker assisted walking with 15-degree slope(III) were significantly different from non-assisted walking (I) (P<0.01). Regarding spatial parameters (velocity and distance), velocity showed the same results as those of temporal parameters (P<0.01); however, distance showed no significant difference among the three conditions (P>0.05). In case of Kinematic parameters (Hip, knee and ankle angle), gait cycle was analyzed by segmenting 6 parameters on the basis of temporal parameter. The Hip joint angle show significant differences in the stance phase (initial double support, initial single support, terminal double support). The knee and ankle joint angle showed significant differences in swing phase (terminal single support). The GA value showed significant differences in each parameters with 3 conditions: (I) non-assisted walking, (II) Rom-walker assisted walking, (III) Rom-walker assisted walking with 15 degree (P<0.05). In this research, in order to verify whether the symmetry of gait is identified, the height of heel and the angle of ankle are restricted during experiment. As the result, the joint angles and velocities of all direction are different during asymmetric gait. According to the results of this experiment using spatio-temporal parameters including kinematic information based on gait phase extracted from AHRS, gait symmetry assessment can be used with the able-bodied. Not only providing real-time data about this and but also figuring out the functional meaning and cause of asymmetry in each phase enable to identify the result which was unknown in previous researches. Additionally this research compared kinematic parameters that extracted from normal subject by confronting them to ordinary walking with those from Scoliosis patients. In case of kinematic parameters, Gait cycle is divided into 6 gait phases with temporal parameters. 20 Healthy subjects that are aged from 20 to 30 and 4 scoliosis subjects are analyzed 5 times during experiment with non-assisted walking conditions. With the analysis result of non-parametric statistical test, GA values that are computed from hip, knee and ankle joint angle showed meaning-full difference between two groups (P<0.05).