MCSNet: Channel Synergy-Based Human-Exoskeleton Interface With Surface Electromyogram

The human–robot interface (HRI) based on biological signals can realize the natural interaction between human and robot. It has been widely used in exoskeleton robots recently to help predict the wearer's movement. Surface electromyography (sEMG)-based HRI has mature applications on the exoskeleton. However, the sEMG signals of paraplegic patients' lower limbs are weak, which means that most HRI based on lower limb sEMG signals cannot be applied to the exoskeleton. Few studies have explored the possibility of using upper limb sEMG signals to predict lower limb movement. In addition, most HRIs do not consider the contribution and synergy of sEMG signal channels. This paper proposes a human–exoskeleton interface based on upper limb sEMG signals to predict lower limb movements of paraplegic patients. The interface constructs an channel synergy-based network (MCSNet) to extract the contribution and synergy of different feature channels. An sEMG data acquisition experiment is designed to verify the effectiveness of MCSNet. The experimental results show that our method has a good movement prediction performance in both within-subject and cross-subject situations, reaching an accuracy of 94.51 and 80.75%, respectively. Furthermore, feature visualization and model ablation analysis show that the features extracted by MCSNet are physiologically interpretable.

Assessing Movement Quality in Youth Footballers: The Relationship between Hip and Lower Limb Movement Screen and Functional Movement Screen

The Hip and Lower Limb Movement Screen (HLLMS) was developed to detect altered movement patterns and asymmetry specifically related to hip, pelvic, and lower limb movement control, as the other tools, such as the Functional Movement Screen (FMS), lacked focus on the hip and pelvic area. Both screening tools contain symmetrical and asymmetrical motor tasks which are based on observation of different aspects of each task performance. One motor task is in both screening tools. Therefore, they have some common features. The present study aimed to assess the relationship between the HLLMS and FMS performance in youth football players. The study included 41 elite male football (soccer) players (age: 15.6 ± 0.50 years), and the HLLMS and FMS scores were analyzed by assessing Spearman’s rank correlation. The FMS total score and the FMSMOVE were moderately correlated with the HLLMS total score (R = −0.54; −0.53, respectively). The FMS rotatory stability task was moderately correlated with the HLLMS small knee bend with the trunk rotation task (R = −0.50). The FMS deep squat task was moderately correlated with the HLLMS deep squat task (R = −0.46). The FMS hurdle step was weakly correlated with two of the HLLMS tasks: standing hip flexion (R = −0.37) and hip abduction with external rotation (R = −0.34). There were no other relationships found (p > 0.05). Out of the seven FMS tasks, only one asymmetrical (trunk rotary stability) and one symmetrical (deep squat) task were moderately related to the newly developed HLLMS tool contributing moderate relationship between the FMS total score and the HLLMS total score. Other FMS tasks were weakly or unrelated with the HLLMS. These findings indicate that these two screening tools mainly assess different aspects of movement quality in healthy youth football players.

Lower limb movement intention recognition for rehabilitation robot aided with projected recurrent neural network

For the lower limb rehabilitation robot, how to better realize intention recognition is the key issue in the practical application. Recognition of the patient’s movement intention is a challenging research work, which needs to be studied from the shallow to the deep. Specifically, it is necessary to ensure that the movement intention of the normal person can be accurately recognized, and then improve the model to realize the recognition of the movement intention of the patients. Therefore, before studying the patient’s movement intention, it is essential to consider the normal person first, which is also for safety considerations. In recent years, a new Hill-based muscle model has been demonstrated to be capable of directly estimating the joint angle intention in an open-loop form. On this basis, by introducing a recurrent neural network (RNN), the whole prediction process can achieve more accuracy in a closed-loop form. However, for the traditional RNN algorithms, the activation function must be convex, which brings some limitations to the solution of practical problems. Especially, when the convergence speed of the traditional RNN model is limited in the practical applications, as the error continues to decrease, the convergence performance of the traditional RNN model will be greatly affected. To this end, a projected recurrent neural network (PRNN) model is proposed, which relaxes the condition of the convex function and can be used in the saturation constraint case. In addition, the corresponding theoretical proof is given, and the PRNN method with saturation constraint has been successfully applied in the experiment of intention recognition of lower limb movement compared with the traditional RNN model.

Ipsilateral Motor Pathways and Transcallosal Inhibition During Lower Limb Movement After Stroke

Background Stroke rehabilitation may be improved with a better understanding of the contribution of ipsilateral motor pathways to the paretic limb and alterations in transcallosal inhibition. Few studies have evaluated these factors during dynamic, bilateral lower limb movements, and it is unclear whether they relate to functional outcomes. Objective Determine if lower limb ipsilateral excitability and transcallosal inhibition after stroke depend on target limb, task, or number of limbs involved, and whether these factors are related to clinical measures. Methods In 29 individuals with stroke, ipsilateral and contralateral responses to transcranial magnetic stimulation were measured in the paretic and nonparetic tibialis anterior during dynamic (unilateral or bilateral ankle dorsiflexion/plantarflexion) and isometric (unilateral dorsiflexion) conditions. Relative ipsilateral excitability and transcallosal inhibition were assessed. Fugl-Meyer, ankle movement accuracy, and walking characteristics were assessed. Results Relative ipsilateral excitability was greater during dynamic than isometric conditions in the paretic limb ( P ≤ .02) and greater in the paretic than the nonparetic limb during dynamic conditions ( P ≤ .004). Transcallosal inhibition was greater in the ipsilesional than contralesional hemisphere ( P = .002) and during dynamic than isometric conditions ( P = .03). Greater ipsilesional transcallosal inhibition was correlated with better ankle movement accuracy ( R2 = 0.18, P = .04). Greater contralateral excitability to the nonparetic limb was correlated with improved walking symmetry ( R2 = 0.19, P = .03). Conclusions Ipsilateral pathways have increased excitability to the paretic limb, particularly during dynamic tasks. Transcallosal inhibition is greater in the ipsilesional than contralesional hemisphere and during dynamic than isometric tasks. Ipsilateral pathways and transcallosal inhibition may influence walking asymmetry and ankle movement accuracy.