Ankle Joint Torque Prediction Using an NMS Solver Informed-ANN Model and Transfer Learning

Abstract

In this work, we predicted ankle joint torque by combining a neuromusculoskeletal (NMS) solver-informed artificial neural network (hybrid-ANN) model with transfer learning based on joint angle and muscle electromyography signals. The hybrid-ANN is an ANN augmented with two kinds of features:(1) experimental measurements -muscle signals and joint angles, and (2) informative physical features extracted from the underlying NMS solver, such as individual muscle force and joint torque. The hybrid-ANN model accuracy in torque prediction was studied in both intra-and inter-subject tests, and compared to the baseline models (NMS and standard-ANN). For each prediction model, seven different cases were studied using data from gait at different speeds and from isokinetic ankle dorsi/plantarflexion motion. Additionally, we integrated a transfer learning method in inter-subject models to improve joint torque prediction accuracy by transferring the learned knowledge from previous participants to a new participant, which could be useful when training data is limited. Our results indicated that better accuracy could be obtained by integrating informative NMS features into a standard ANN model, especially in inter-subject cases; overall, the hybrid-ANN model predicted joint torque with higher accuracy than the baseline models, most notably in inter-subject prediction after adopting the transfer learning technique. We demonstrated the potential of combining physics-based NMS and standard-ANN models with a transfer learning technique in different prediction scenarios. This procedure holds great promise in applications such as assistance-as-needed exoskeleton control strategy design by incorporating the physiological joint torque of the users.

Advanced Methods and Applications for Neurointelligence

“…This is likely due to the physical characteristics adopted from NMS models, which show a noticeable offset at the start of each movement. In NMS models, two prior time steps of neural activation from each MTU are required to calculate muscle neural activation (Zhang et al, 2022). At the beginning of a cycle, these past two neural activation values are not yet obtainable and are approximated using EMG signals from two previous time steps, potentially leading to initial inaccuracies in predicted torque.…”

“…To improve prediction accuracy, ANNs have been integrated into NMS models in recent research. In our recent study (Zhang et al, 2022), an NMS solver-informed ANN model was developed to estimate ankle joint torque by combining features from an NMS model with a standard ANN, based on measured joint angles and muscle EMG signals during gait and isokinetic motions. This hybrid model was overall more accurate than the NMS or standard ANN models alone, but still showed poor prediction performance in one subject during gait, possibly due to incorporating less informative or misleading input features from the NMS model.…”

Estimation of Joint Torque by EMG-Driven Neuromusculoskeletal Models and LSTM Networks

“…As such, for applications in which participant-specific muscle-tendon parameters and relationships are desired, an NMS model-based solution is clearly more suitable. A model that combines relationships defined from NMS models with neural network structures may have the advantages of higher prediction accuracy than either of its derivatives, in addition to the benefit of defined muscletendon properties and relationships [7], [33], [34]. Therefore, in future work, it would be worthwhile to explore the potential of a physics-informed neural network approach that combines NMS features with an LSTM model to enhance the neural network method's extrapolation capability.…”

Neuromusculoskeletal model-informed machine learning-based control of a knee exoskeleton with uncertainties quantification

“…This is likely due to the physical characteristics adopted from NMS models, which show a noticeable offset at the start of each movement. In NMS models, two prior time steps of neural activation from each MTU are required to calculate muscle neural activation (Zhang et al, 2022 ). At the beginning of a cycle, these past two neural activation values are not yet obtainable and are approximated using EMG signals from two previous time steps, potentially leading to initial inaccuracies in predicted torque.…”

“…To improve prediction accuracy, ANNs have been integrated into NMS models in recent research. In our recent study (Zhang et al, 2022 ), an NMS solver-informed ANN model was developed to estimate ankle joint torque by combining features from an NMS model with a standard ANN, based on measured joint angles and muscle EMG signals during gait and isokinetic motions. This hybrid model was overall more accurate than the NMS or standard ANN models alone, but still showed poor prediction performance in one subject during gait, possibly due to incorporating less informative or misleading input features from the NMS model.…”