Asymmetric Gait Analysis Using a DTW Algorithm with Combined Gyroscope and Pressure Sensor

Walking is one of the most basic human activities. Various diseases may be caused by abnormal walking, and abnormal walking is mostly caused by disease. There are various characteristics of abnormal walking, but in general, it can be judged as asymmetric walking. Generally, spatiotemporal parameters can be used to determine asymmetric walking. The spatiotemporal parameter has the disadvantage that it does not consider the influence of the diversity of patterns and the walking speed. Therefore, in this paper, we propose a method to analyze asymmetric walking using Dynamic Time Warping (DTW) distance, a time series analysis method. The DTW distance was obtained by combining gyroscope data and pressure data. The experiment was carried out by performing symmetrical walking and asymmetrical walking, and asymmetric walking was performed as a simulation of hemiplegic walking by fixing one ankle using an auxiliary device. The proposed method was compared with the existing asymmetric gait analysis method. As a result of the experiment, a p-value lower than 0.05 was obtained, which proved that there was a statistically significant difference.

Walking on a Vertically Oscillating Platform with Simulated Gait Asymmetry

Asymmetric gait is associated with pain, injury, and reduced stability in patient populations. Data from side by side walking suggest that unintentional synchronization with an external cue may reduce gait asymmetry. Two types of asymmetric gait were examined here: (1) mass imbalance between limbs to simulate single limb amputation and (2) restriction of plantarflexion during toe-off to simulate reduced propulsion from neurological impairment. Twenty-five healthy participants walked normally and with simulated gait asymmetry on a custom-designed treadmill that oscillated in the vertical direction via pneumatic actuation (amplitude: 2 cm, frequency: participant’s preferred step frequency). Swing Time Asymmetry (STA) and Phase Coordination Index (PCI) both increased significantly with the application of unilateral mass and plantarflexion restriction (p < 0.001). However, walking with simulated asymmetry did not alter unintentional synchronization with the treadmill motion. Further, oscillation of the treadmill did not improve STA or PCI while walking with simulated asymmetry. Analysis of synchronized step clusters using the Weibull survival function revealed that synchronization with the platform persisted for longer durations when compared with data from side by side walking. These results suggest that walking on a vertically oscillating surface may not be an effective approach for improving gait asymmetry.

Novel velocity estimation for symmetric and asymmetric self-paced treadmill training

Background Self-paced treadmills (SPT) can provide an engaging setting for gait rehabilitation by responding directly to the user’s intent to modulate the external environment and internal effort. They also can improve gait analyses by allowing scientists and clinicians to directly measure the effect of an intervention on walking velocity. Unfortunately, many common SPT algorithms are not suitable for individuals with gait impairment because they are designed for symmetric gait patterns. When the user’s gait is asymmetric due to paresis or if it contains large accelerations, the performance is diminished. Creating and validating an SPT that is suitable for asymmetric gait will improve our ability to study rehabilitation interventions in populations with gait impairment. The objective of this study was to test and validate a novel self-paced treadmill on both symmetric and asymmetric gait patterns and evaluate differences in gait kinematics, kinetics, and muscle activity between fixed-speed and self-paced treadmill walking. Methods We collected motion capture, ground reaction force data, and muscle activity from 6 muscles in the dominant leg during walking from 8 unimpaired subjects. In the baseline condition, the subjects walked at 3 fixed-speeds normalized to their leg length as Froude numbers. We developed a novel kinematic method for increasing the accuracy of the user’s estimated walking velocity and compared our method against other published algorithms at each speed. Afterward, subjects walked on the SPT while matching their walking speed to a given target velocity using visual feedback of the treadmill speed. We evaluated the SPT by measuring steady-state error and the number of steps to reach the desired speed. We split the gait cycle into 7 phases and compared the kinematic, kinetic, and muscle activity between the fixed speed and self-paced mode in each phase. Then, we validated the performance of the SPT for asymmetric gait by having subjects walk on the SPT while wearing a locked-knee brace set to 0° on the non-dominant leg. Results Our SPT enabled controlled walking for both symmetric and asymmetric gait patterns. Starting from rest, subjects were able to control the SPT to reach the targeted speeds using visual feedback in 13–21 steps. With the locked knee brace, subjects controlled the treadmill with substantial step length and step velocity asymmetry. One subject was able to execute a step-to gait and halt the treadmill on heel-strikes with the braced leg. Our kinematic correction for step-length outperformed the competing algorithms by significantly reducing the velocity estimation error at the tested velocities. The joint kinematics, joint torques, and muscle activity were generally similar between fixed-speed and self-paced walking. Statistically significant differences were found in 5 of 63 tests for joint kinematics, 2 of 63 tests for joint torques, and 9 of 126 tests for muscle activity. The differences that were statistically significant were not found across all speeds and were generally small enough to be of limited clinical relevance. Conclusions We present a validated method for implementing a self-paced treadmill for asymmetric and symmetric gaits. As a result of the increased accuracy of our estimation algorithm, our SPT produced controlled walking without including a position feedback controller, thereby reducing the influence of the controller on measurements of the user’s true walking speed. Our method relies only on a kinematic correction to step length and step time which can support transfer to systems outside of the laboratory for symmetric and asymmetric gaits in clinical populations.