The Classification of Minor Gait Alterations Using Wearable Sensors and Deep Learning

Abstract

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.

Non-Linear Template-Based Approach for the Study of Locomotion

“…The use of machine learning algorithms has shown its usefulness in the search for an individual gait pattern and its evolution over different time scales [28]. This gait pattern recognition can be used to successfully solve tasks for classifying gait disorders [29][30][31][32] or for extracting gait characteristics [33]. Deep learning methods have also been used to characterise the gait phase the subject is in, thus resulting in IC/FC detection from multiple accelerometers [34], 3D markers [35,36] or instrumented shoes [37].…”

Supporting the Assessment of Hereditary Transthyretin Amyloidosis Patients Based On 3-D Gait Analysis and Machine Learning

“…In the context of movement-related diseases, ML/DL techniques have been used, together with data provided by wearable or vision-based sensors, to support gait assessment with the aim of diagnosis and/or evaluation of disease progression [10,11,[45][46][47][48][49][50][51][52][53][54][55][56][57]. The main focus of most contributions is the detection of abnormal gait based on information extracted from gait data obtained with accelerometers, gyroscopes and/or pressure sensors [45,46,56,57], or with RGB-D cameras [10,[47][48][49][50][51][52][53][54][55].…”

Detection of Gait Abnormalities for Fall Risk Assessment Using Wrist-Worn Inertial Sensors and Deep Learning

“…Recent technological progress has led to the development of various devices that allow for continuous human gait monitoring in free-living environments. Current sensor-based gait analysis systems are designed using either external sensors, such as cameras [ 8 ] and pressure sensors [ 9 , 10 ], or wearable sensors. However, because of the ability of the wearable sensors to provide a reliable insight into an individual’s gait quality in the least obtrusive way, they are becoming the most attractive approach for gait analysis and fall risk assessment.…”

“…They explored bidirectional LSTM networks that incorporated sequences of spatiotemporal gait parameters, as well as raw inertial data, to classify high fall risk and low fall risk patients. Similarly, in [ 9 ], LSTM networks were employed to classify artificially induced gait alterations from sensors worn inside the shoes. All of these studies show the potential of DL methods for fall risk assessment using wearable sensor data, achieving an accuracy of 76–82%.…”