Influence of pre-processing in the extraction of muscle synergies during human locomotion

Long short-term memory (LSTM) recurrent neural network for muscle activity detection

BACKGROUND

The accurate temporal analysis of muscle activation is of great interest in many research areas, spanning from neurorobotic systems to the assessment of altered locomotion patterns in orthopedic and neurological patients and the monitoring of their motor rehabilitation. The performance of the existing muscle activity detectors is strongly affected by both the SNR of the surface electromyography (sEMG) signals and the set of features used to detect the activation intervals. This work aims at introducing and validating a powerful approach to detect muscle activation intervals from sEMG signals, based on long short-term memory (LSTM) recurrent neural networks.

METHODS

First, the applicability of the proposed LSTM-based muscle activity detector (LSTM-MAD) is studied through simulated sEMG signals, comparing the LSTM-MAD performance against other two widely used approaches, i.e., the standard approach based on Teager-Kaiser Energy Operator (TKEO) and the traditional approach, used in clinical gait analysis, based on a double-threshold statistical detector (Stat). Second, the effect of the Signal-to-Noise Ratio (SNR) on the performance of the LSTM-MAD is assessed considering simulated signals with nine different SNR values. Finally, the newly introduced approach is validated on real sEMG signals, acquired during both physiological and pathological gait. Electromyography recordings from a total of 20 subjects (8 healthy individuals, 6 orthopedic patients, and 6 neurological patients) were included in the analysis.

RESULTS

The proposed algorithm overcomes the main limitations of the other tested approaches and it works directly on sEMG signals, without the need for background-noise and SNR estimation (as in Stat). Results demonstrate that LSTM-MAD outperforms the other approaches, revealing higher values of F1-score (F1-score > 0.91) and Jaccard similarity index (Jaccard > 0.85), and lower values of onset/offset bias (average absolute bias < 6 ms), both on simulated and real sEMG signals. Moreover, the advantages of using the LSTM-MAD algorithm are particularly evident for signals featuring a low to medium SNR.

CONCLUSIONS

The presented approach LSTM-MAD revealed excellent performances against TKEO and Stat. The validation carried out both on simulated and real signals, considering normal as well as pathological motor function during locomotion, demonstrated that it can be considered a powerful tool in the accurate and effective recognition/distinction of muscle activity from background noise in sEMG signals.

Surface Electromyography Applied to Gait Analysis: How to Improve Its Impact in Clinics?

Surface electromyography (sEMG) is the main non-invasive tool used to record the electrical activity of muscles during dynamic tasks. In clinical gait analysis, a number of techniques have been developed to obtain and interpret the muscle activation patterns of patients showing altered locomotion. However, the body of knowledge described in these studies is very seldom translated into routine clinical practice. The aim of this work is to analyze critically the key factors limiting the extensive use of these powerful techniques among clinicians. A thorough understanding of these limiting factors will provide an important opportunity to overcome limitations through specific actions, and advance toward an evidence-based approach to rehabilitation based on objective findings and measurements.

Muscle Synergies Extracted Using Principal Activations: Improvement of Robustness and Interpretability

The muscle synergy theory has been widely used to assess the modular organization of the central nervous system (CNS) during human locomotion. The pre-processing approach applied to the surface electromyographic (sEMG) signals influences the extraction of muscle synergies. The aim of this contribution is to assess the improvements in muscle synergy extraction obtained by using an innovative pre-processing approach. We evaluate the improvement in terms of the possible variation in the number of muscle synergies, of the intra-subject consistency, of the robustness, and of the interpretability of the results. The pre-processing approach presented in this paper is based on the extraction of the muscle principal activations (muscle activations strictly necessary to accomplish a specific biomechanical task) from the original sEMG signals, to then obtain muscle synergies using principal activations only. The results herein presented show that the application of this novel approach for the extraction of the muscle synergies provides a more robust and easily interpretable description of the modular organization of the CNS with respect to the standard pre-processing approach.

Asymmetry Index in Muscle Activations

Gait asymmetry is typically evaluated using spatio-temporal or joint kinematics parameters. Only a few studies addressed the problem of defining an asymmetry index directly based on muscle activity, extracting parameters from surface electromyography (sEMG) signals. Moreover, no studies used the extraction of the muscle principal activations (activations that are necessary for accomplishing a specific motor task) as the base to construct an asymmetry index, less affected by the variability of sEMG patterns. The aim of this paper is to define a robust index to quantitatively assess the asymmetry of muscle activations during locomotion, based on the extraction of the principal activations. SEMG signals were analyzed combining statistical gait analysis (SGA) and a clustering algorithm that allows for obtaining the muscle principal activations. We evaluated the asymmetry levels of four lower limb muscles in: (1) healthy subjects of different ages (children, adults, and elderly); (2) different populations of orthopedic patients (adults with megaprosthesis of the knee after bone tumor resection, elderly subjects after total knee arthroplasty, and elderly subjects after total hip arthroplasty); and (3) neurological patients (children with hemiplegic cerebral palsy and elderly subjects affected by idiopathic normal pressure hydrocephalus). The asymmetry index obtained for each pathological population was then compared to that of age-matched controls. We found asymmetry levels consistent with the expected impact of the different pathologies on muscle activation during gait. This suggests that the proposed index can be successfully used in clinics for an objective assessment of the muscle activation asymmetry during locomotion.