3 March 2022 in IEEE Transactions on Neural Systems and Rehabilitation Engineering
Estimation of joint torque during movement provides important information in several settings, such as effect of athletes’ training or of a medical intervention, or analysis of the remaining muscle strength in a wearer of an assistive device. The ability to estimate joint torque during daily activities using wearable sensors is increasingly relevant in such settings. In this study, lower limb joint torques during ten daily activities were predicted by long short-term memory (LSTM) neural networks and transfer learning. LSTM models were trained with muscle electromyography signals and lower limb joint angles. Hip flexion/extension, hip abduction/adduction, knee flexion/extension and ankle dorsiflexion/plantarflexion torques were predicted. The LSTM models’ performance in predicting torque was investigated in both intra-subject and inter-subject scenarios. Each scenario was further divided into intra-task and inter-task tests. We observed that LSTM models could predict lower limb joint torques during various activities accurately with relatively low error (root mean square error ≤ 0.14 Nm/kg, normalized root mean square error ≤ 8.7%) either through a uniform model or through ten separate models in intra-subject tests. Furthermore, a transfer learning technique was adopted in the inter-task and inter-subject tests to further improve the generalizability of LSTM models by pre-training a model on multiple subjects and/or tasks and transferring the learned knowledge to a target task/subject. Particularly in the inter-subject tests, we could predict joint torques accurately in several movements after training from only a few movements from new subjects.
1 April 2021 in IEEE Transactions on Automation Science and Engineering
EFElena M. Gutierrez Farewik
Elena M. Gutierrez Farewik
In recent decades, there has been an increasing interest in the use of robotic powered exoskeletons to assist patients with movement disorders in rehabilitation and daily life. Providing assistive torque that compensates for the user's remaining muscle contributions is a growing and challenging field within exoskeleton control. In this article, ankle joint torques were estimated using electromyography (EMG)-driven neuromusculoskeletal (NMS) model and an artificial neural network (ANN) model in seven movement tasks, including fast walking, slow walking, self-selected speed walking, and isokinetic dorsi/plantar flexion at 60°/s and 90°/s. In each method, EMG signals and ankle joint angles were used as input, the models were trained with data from 3-D motion analysis, and ankle joint torques were predicted. Six cases using different motion trials as calibration (for the NMS model)/training (for the ANN) were devised, and the agreement between the predicted and measured ankle joint torques was computed. We found that the NMS model could overall better predict ankle joint torques from EMG and angle data than the ANN model with some exceptions; the ANN predicted ankle joint torques with better agreement when trained with data from the same movement. The NMS model predicted ankle joint torque best when calibrated with trials during which EMG reached maximum levels, whereas the ANN predicted well when trained with many trials and types of movements. In addition, the ANN prediction may become less reliable when predicting unseen movements. Detailed comparative studies of methods to predict ankle joint torque are crucial for determining strategies for exoskeleton control.
7 June 2021 in IEEE Transactions on Neural Systems and Rehabilitation Engineering
Detecting human movement intentions is fundamental to neural control of robotic exoskeletons, as it is essential for achieving seamless transitions between different locomotion modes. In this study, we enhanced a muscle synergy-inspired method of locomotion mode identification by fusing the electromyography data with two types of data from wearable sensors (inertial measurement units), namely linear acceleration and angular velocity. From the finite state machine perspective, the enhanced method was used to systematically identify 2 static modes, 7 dynamic modes, and 27 transitions among them. In addition to the five broadly studied modes (level ground walking, ramps ascent/descent, stairs ascent/descent), we identified the transition between different walking speeds and modes of ramp walking at different inclination angles. Seven combinations of sensor fusion were conducted, on experimental data from 8 able-bodied adult subjects, and their classification accuracy and prediction time were compared. Prediction based on a fusion of electromyography and gyroscope (angular velocity) data predicted transitions earlier and with higher accuracy. All transitions and modes were identified with a total average classification accuracy of 94.5% with fused sensor data. For nearly all transitions, we were able to predict the next locomotion mode 300-500ms prior to the step into that mode.
14 September 2023 in IEEE Transactions on Neural Systems and Rehabilitation Engineering
Accurately predicting joint torque using wearable sensors is crucial for designing assist-as-needed exoskeleton controllers to assist muscle-generated torque and ensure successful task performance. In this paper, we estimated ankle dorsiflexion/plantarflexion, knee flexion/extension, hip flexion/extension, and hip abduction/adduction torques from electromyography (EMG) and kinematics during daily activities using neuromusculoskeletal (NMS) models and long short-term memory (LSTM) networks. The joint torque ground truth for model calibrating and training was obtained through inverse dynamics of captured motion data. A cluster approach that grouped movements based on characteristic similarity was implemented, and its ability to improve the estimation accuracy of both NMS and LSTM models was evaluated. We compared torque estimation accuracy of NMS and LSTM models in three cases: Pooled, Individual, and Clustered models. Pooled models used data from all 10 movements to calibrate or train one model, Individual models used data from each individual movement, and Clustered models used data from each cluster. Individual, Clustered and Pooled LSTM models all had relatively high joint torque estimation accuracy. Individual and Clustered NMS models had similarly good estimation performance whereas the Pooled model may be too generic to satisfy all movement patterns. While the cluster approach improved the estimation accuracy in NMS models in some movements, it made relatively little difference in the LSTM neural networks, which already had high estimation accuracy. Our study provides practical implications for designing assist-as-needed exoskeleton controllers by offering guidelines for selecting the appropriate model for different scenarios, and has potential to enhance the functionality of wearable exoskeletons and improve rehabilitation and assistance for individuals with motor disorders.
8 July 2017 in Medical Engineering & Physics
Quantifying neural and non-neural contributions to increased joint resistance in spasticity is essential for a better understanding of its pathophysiological mechanisms and evaluating different intervention strategies. However, direct measurement of spasticity-related manifestations, e.g., motoneuron and biophysical properties in humans, is extremely challenging. In this vein, we developed a forward neuromusculoskeletal model that accounts for dynamics of muscle spindles, motoneuron pools, muscle activation and musculotendon of wrist flexors and relies on the joint angle and resistant torque as the only input measurement variables. By modeling the stretch reflex pathway, neural and non-neural related properties of the spastic wrist flexors were estimated during the wrist extension test. Joint angle and resistant torque were collected from 17 persons with chronic stroke and healthy controls using NeuroFlexor, a motorized force measurement device during the passive wrist extension test. The model was optimized by tuning the passive and stretch reflex-related parameters to fit the measured torque in each participant. We found that persons with moderate and severe spasticity had significantly higher stiffness than controls. Among subgroups of stroke survivors, the increased neural component was mainly due to a lower muscle spindle rate at 50% of the motoneuron recruitment. The motoneuron pool threshold was highly correlated to the motoneuron pool gain in all subgroups. The model can describe the overall resistant behavior of the wrist joint during the test. Compared to controls, increased resistance was predominantly due to higher elasticity and neural components. We concluded that in combination with the NeuroFlexor measurement, the proposed neuromusculoskeletal model and optimization scheme served as suitable tools for investigating potential parameter changes along the stretch-reflex pathway in persons with spasticity.
16 September 2022 in IEEE Journal of Biomedical and Health Informatics
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.