Accelerometry-enabled measurement of walking performance with a robotic exoskeleton: a pilot study

Abstract

BackgroundClinical scores for evaluating walking skills with lower limb exoskeletons are often based on a single variable, such as distance walked or speed, even in cases where a host of features are measured. We investigated how to combine multiple features such that the resulting score has high discriminatory power, in particular with few patients. A new score is introduced that allows quantifying the walking ability of patients with spinal cord injury when using a powered exoskeleton.MethodsFour spinal cord injury patients were trained to walk over ground with the ReWalk™ exoskeleton. Body accelerations during use of the device were recorded by a wearable accelerometer and 4 features to evaluate walking skills were computed. The new score is the Gaussian naïve Bayes surprise, which evaluates patients relative to the features’ distribution measured in 7 expert users of the ReWalk™. We compared our score based on all the features with a standard outcome measure, which is based on number of steps only.ResultsAll 4 patients improved over the course of training, as their scores trended towards the expert users’ scores. The combined score (Gaussian naïve surprise) was considerably more discriminative than the one using only walked distance (steps). At the end of training, 3 out of 4 patients were significantly different from the experts, according to the combined score (p < .001, Wilcoxon Signed-Rank Test). In contrast, all but one patient were scored as experts when number of steps was the only feature.ConclusionIntegrating multiple features could provide a more robust metric to measure patients’ skills while they learn to walk with a robotic exoskeleton. Testing this approach with other features and more subjects remains as future work.

Effectiveness of over-ground robotic locomotor training in improving walking performance, cardiovascular demands, secondary complications and user-satisfaction in individuals with spinal cord injuries: A systematic review

“…Six studies considered the mean distance and velocity achieved during a 6MWT showing a range from 47 to 129 m and 0.22 to 0.36 m/s, respectively (6, 17-20, 22) (Appendix I). Six studies considered the velocity required to complete a 10MWT (Appendix II), ranging from 0.25 to 0.38 m/s across 4 studies (17,18,20,21). The remaining 2 studies indicated that different injury levels can affect walking velocity (22), as can the level of assistance provided while walking (23).…”

Computer vision to automatically assess infant neuromotor risk

“…We estimated the probability of each infant belonging to the healthy reference population, represented by the YouTube cohort in this case, who are presumed to be healthy. We adopted the Naïve Bayes approach, previously used for clinical assessments of walking performance (Lonini et al, 2016). Under the assumption of normally-distributed features and feature independence, the joint probability of features over the reference population is:…”

“…infant deviates from the typical movements of healthy infants using a single score, the Naïve Gaussian Bayesian Surprise (Lonini et al, 2016). When we then tested this system on a clinical population (N=19) where the level of neuromotor risk was assessed by a clinician (low, moderate, and high), we found that Bayesian Surprise varied across participant groups.…”

Test methods for exoskeletons—lessons learned from industrial and response robotics

“…Additional measurements that are not explicitly discussed here, but that can be conducted in conjunction with the proposed tests include clinical measures of walking performance [8] and other measures of human performance, such as might be used in the rehabilitation community. Furthermore, there are measures of vitals, including heart rate, blood pressure, oxygen demand, etc.…”