This paper describes how non-invasive wearable sensors can be used in combination with deep learning to classify artificially induced gait alterations without the requirement for a medical professional or gait analyst to be present. This approach is motivated by the goal of diagnosing gait abnormalities on a symptom by symptom basis, irrespective of other neuromuscular movement disorders patients may be affected by. This could lead to improvements in treatment and offer a greater insight into movement disorders. Methods: Inshoe pressure was measured for 12 able-bodied participants, each subject to 8 artificially induced gait alterations, achieved by modifying the underside of the shoe. The data were recorded at 100 Hz over 2520 data channels and was analysed using the deep learning architecture, long term short term memory networks. Additionally, the rationale for the decision-making process of these networks was investigated. Conclusion: Long term short term memory networks are applicable to the classification of gait function. The classifications can be made using only 2 seconds of sparse data (82.0% accuracy over 96,000 instances of test data) from participants who were not part of the training set. Significance: This work provides potential for gait function to be accurately classified using non-invasive techniques, and at more regular intervals, outside of a clinical setting without the need for healthcare professionals to be present.

No abstract

No abstract

The use of RAGT in SCI individuals has the potential to benefit upright locomotion of SCI individuals. Its use should not replace other therapies but be incorporated into a multi-modality rehabilitation approach.

BackgroundOverground lower-limb robotic exoskeletons are assistive devices used to facilitate ambulation and gait rehabilitation. Our understanding of how closely they resemble comfortable and slow walking is limited. This information is important to maximise the effects of gait rehabilitation. The aim was to compare the 3D gait parameters of able-bodied individuals walking with and without an exoskeleton at two speeds (self-selected comfortable vs. slow, speed-matched to the exoskeleton) to understand how the user's body moved within the device. MethodsEight healthy, able-bodied individuals walked along a 12-metre walkway with and without the exoskeleton. Three-dimensional whole-body kinematics inside the device were captured. Temporal-spatial parameters and sagittal joint kinematics were determined for normal and exoskeleton walking. One-way repeated measures ANOVAs and statistical parametric mapping were used to compare the three walking conditions ( P <0.05). FindingsThe walking speeds of the slow (0.44[0.03] m/s) and exoskeleton (0.41[0.03] m/s) conditions were significantly slower than the comfortable walking speed (1.54[0.07] m/s). However, time in swing was significantly greater ( P <0.001, d =-3.64) and double support was correspondingly lower ( P <0.001, d =3.72) during exoskeleton gait than slow walking, more closely resembling comfortable speed walking. Ankle and knee angles were significantly reduced in the slow and exoskeleton conditions. Angles were also significantly different for the upper body. InterpretationAlthough the slow condition was speed-matched to exoskeleton gait, the stance:swing ratio of exoskeleton stepping more closely resembled comfortable gait than slow gait. The altered upper body kinematics suggested that overground exoskeletons may provide a training environment that would also benefit balance training.