Gait event detection using linear accelerometers or angular velocity transducers in able-bodied and spinal-cord injured individuals

Abstract

We report on three different methods of gait event detection (toe-off and heel strike) using miniature linear accelerometers and angular velocity transducers in comparison to using standard pressure-sensitive foot switches. Detection was performed with normal and spinal-cord injured subjects. The detection of end contact (EC), normally toe-off, and initial contact (IC) normally, heel strike was based on either foot linear accelerations or foot sagittal angular velocity or shank sagittal angular velocity. The results showed that all three methods were as accurate as foot switches in estimating times of IC and EC for normal gait patterns. In spinal-cord injured subjects, shank angular velocity was significantly less accurate (p<0.02). We conclude that detection based on foot linear accelerations or foot angular velocity can correctly identify the timing of IC and EC events in both normal and spinal-cord injured subjects.

A Real-Time Gait Event Detection for Lower Limb Prosthesis Control and Evaluation

“…However, they are susceptible to mechanical failure, have generally poor durability and low cosmetic acceptance [7]. Among wearable sensors, accelerometers and gyroscopes are being deployed at different body locations for long-term monitoring of human gait [3,[6][7][8][9][10][11][12][13]. Recent advancements allow these sensors to be miniaturized, with faster processing capability and higher memory capacities to support outdoor applications [9].…”

Gait Event Detection in Real-World Environment for Long-Term Applications: Incorporating Domain Knowledge Into Time-Frequency Analysis

“…In the context of gyro-based algorithms, many methods have been developed from angular velocity signals obtained from shank-attached gyroscopes. For example, the approach in [18] uses adaptive thresholds while [13], [19] use peak detection to identify HS and TO from angular velocity signals. Other approaches include [20], where the gait cycle is divided into four gait phases represented in the form of a state machine and the transitions are governed by a knowledgebased algorithm, and [21], where an online Hidden Markov Model based method is presented.…”

“…On the other hand, accelerometers are also being increasingly used as they are low powered devices, in the range of few microamperes, and have been shown to provide reliable measures of gait parameters [17], [23]. Most algorithms analyze signals obtained from individual accelerometer axis by positioning the sensor in a specific pre-defined orientation [2], [3], [13], [24]- [28] with the assumption that the accelerometer shall stay statically positioned throughout the experiment. However, it is quite likely that external factors might disturb the original configuration during long-term analysis [28], and thus either the axis alignment should be checked and readjusted frequently or the exact orientation of the accelerometer must be known throughout, to compensate for the misalignment of the axes.…”

Real-time gait event detection for transfemoral amputees during ramp ascending and descending

“…Objective gait analysis has been employed in multiple applications: evaluation of kinematic and kinetic parameters of human gait, design and optimization of assistive devices [4][5][6], ambulatory monitoring method for applications to Parkinson's disease [7], rehabilitations, prosthetics and orthotics [8,9] . Gait analysis incorporates a variety of parameters used for objective assessment of gait such as cadence, speed, stride, step length, single and double support time percentage [10]. The temporal properties of gait and evaluation of gait trajectories are entirely based on accurate detection of the key events such as initial contact (IC) and toe off (TO) for segmentation of gait cycle.…”