Predicting biological joint moment during multiple ambulation tasks

Camargo, Jonathan; Molinaro, Dean D.; Young, Aaron

doi:10.1016/j.jbiomech.2022.111020

Cited by 26 publications

(27 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Machine learning methods to capture dynamics have been used across physics, engineering, and neuroscience to learn the dynamics underlying complex systems when the governing equations are unknown (Bongard and Lipson, 2007; Pandarinath et al, 2018; Sanchez-Gonzalez et al, 2020). Recently, machine learning models have been used in human gait to predict continuous kinetic variables such as ground-reaction forces (Alcantara et al, 2022) or joint torque (Camargo et al, 2022; Giarmatzis et al, 2020) based on kinematic data. Dynamical machine learning models have also been used to encode gait dynamics, including responses to perturbations or assistive devices, but their model structure did not enable comparisons between individuals (Berrueta et al, 2019; Drnach et al, 2019; Maus et al, 2015; Rosenberg et al, 2020; Wang and Srinivasan, 2013).…”

Section: Discussionmentioning

confidence: 99%

Discovering individual-specific gait signatures from data-driven models of neuromechanical dynamics

Winner

Rosenberg

Kesar

et al. 2022

Preprint

View full text Add to dashboard Cite

Locomotion results from the interactions of highly nonlinear neural and biomechanical dynamics. Accordingly, understanding gait dynamics across behavioral conditions and individuals based on detailed modeling of the underlying neuromechanical system has proven difficult. Here, we develop a data-driven and generative modeling approach that recapitulates the dynamical features of gait behaviors to enable more holistic and interpretable characterizations and comparisons of gait dynamics. Specifically, gait dynamics of multiple individuals are predicted by a dynamical model that defines a common, low-dimensional, latent space to compare group and individual differences. We find that highly individualized dynamics - i.e., gait signatures - for healthy older adults and stroke survivors during treadmill walking are conserved across gait speed. Gait signatures further reveal individual differences in gait dynamics, even in individuals with similar functional deficits. Moreover, components of gait signatures can be biomechanically interpreted and manipulated to reveal their relationships to observed spatiotemporal joint coordination patterns. Lastly, the gait dynamics model can predict the time evolution of joint coordination based on an initial static posture. Our gait signatures framework thus provides a generalizable, holistic method for characterizing and predicting cyclic, dynamical motor behavior that may generalize across species, pathologies, and gait perturbations.

show abstract

Section: Discussionmentioning

confidence: 99%

Discovering individual-specific gait signatures from data-driven models of neuromechanical dynamics

Winner

Rosenberg

Kesar

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…This study showed MLP has better performance in joint moment prediction compared with other models and LSTM would be considered for realtime estimation. More recently, Camargo et al [35] estimated hip, knee, and ankle joint moments in multiple locomotion modes (treadmill, stair, ramp) using extracted features from the cluster of electrogoniometer, EMG, and IMU sensors' data, performed feature selection, and then used the selected feature as input into the Artificial Neural Network (ANN) and XGBoost. However, the total number of 18 sensors and the training set that established from a single subject make doubt on practicality in real-world deployment and reproducibilty with unseen human subjects.…”

Section: Related Workmentioning

confidence: 99%

“…Recently, researchers are focusing on acquiring kinetics parameters through data-driven method [34]- [43], [43]- [46]. However, these studies are still relying on a large number of sensors [35], reversely acquired simulated IMU data from the retrospective motion capture data [34], [37], and highly repetitive treadmill/level-ground walking [36]- [43], [43]- [46]. Since experimentally collected IMU data contains noise induced from multi-frequncy vibration during the impact of the limb on the ground, the simulated IMU data that was acquired from retrospectively captured motion capture data may not provide a reliable kinetics estimation during real application.…”

Section: Introductionmentioning

confidence: 99%

Estimation of Lower Extremity Joint Moments and 3D Ground Reaction Forces Using IMU Sensors in Multiple Walking Conditions: A Deep Learning Approach

Hossain

Guo

Choi

2023

IEEE J. Biomed. Health Inform.

View full text Add to dashboard Cite

Human kinetics, specifically joint moments and ground reaction forces (GRFs) can provide important clinical information and can be used to control assistive devices. Traditionally, collection of kinetics is mostly limited to the lab environment because it relies on data that measured from a motion capture system and floor-embedded force plates to calculate the dynamics via musculoskeletal models. This spatially limited method makes it extremely challenging to measure kinetics outside the laboratory in a variety of walking conditions due to the expensive device setup and large space required. Recently, employing machine learning with IMU sensors are suggested as an alternative method for biomechanical analyses. Although these methods enable estimating human kinetic data outside the laboratory by linking IMU sensor data with kinetics dataset, they were limited to show inaccurate kinetic estimates even in highly repeatable single walking conditions due to the employment of generic deep learning algorithms. Thus, this paper proposes a novel deep learning model, Kinetics-FM-DLR-Ensemble-Net to estimate the hip, knee, and ankle joint moments in the sagittal plane and 3 dimensional ground reaction forces (GRFs) using three IMU sensors on the thigh, shank, and foot under several representative walking conditions in daily living, such as treadmill, level-ground, stair, and ramp with different walking speeds. This is the first study that implements both joint moments and GRFs in multiple walking conditions using IMU sensors via deep learning. Our deep learning model is versatile and accurate for identifying human kinetics across diverse subjects and walking conditions and our model outperforms state-of-the-art deep learning model for kinetics estimation by a large margin.

show abstract

“…More recently, studies have successfully used more advanced neural network classes such as recurrent neural networks (RNNs) [19] and attention-based long short-term memory networks (LSTMs) [20], [21] to generate reference trajectories for prosthesis controllers based on able-bodied data. Others have used wearable sensors and machine learning models to predict joint moments for exoskeleton control, also using ablebodied data [22]. However, deep learning models have not been used to predict prosthesis joint dynamics directly for individuals with transtibial amputation.…”

Section: B Current Data-driven Approaches In Prosthetic Controlmentioning

confidence: 99%

Using Deep Learning Models to Predict Prosthetic Ankle Torque

Prasanna¹,

Realmuto²,

Anderson³

et al. 2022

Preprint

View full text Add to dashboard Cite

<p>Inverse dynamics from motion capture is the most common technique for acquiring biomechanical kinetic data. However, this method is time-intensive, limited to a gait laboratory setting, and requires a large array of reflective markers to be attached to the body. A practical alternative must be developed to provide biomechanical information to high-bandwidth prosthesis control systems to enable predictive controllers. In this study, we applied deep learning to build dynamical system models capable of accurately estimating and predicting prosthetic ankle torque from inverse dynamics using only six input signals. We performed a hyperparameter optimization protocol that automatically selected the model architectures and learning parameters that resulted in the most accurate predictions. We show that the trained deep neural networks predict ankle torques one sample into the future with an average RMSE of 0.04±0.02 Nm/kg, corresponding to 2.9±1.6% of the ankle torque’s dynamic range. Comparatively, a manually-derived analytical regression model predicted ankle torques with a RMSE of 0.35±0.53 Nm/kg, corresponding to 26.6±40.9% of the ankle torque’s dynamic range. In addition, the deep neural networks predicted ankle torque values half a gait cycle into the future with an average drop in performance of 1.7% of the ankle torque’s dynamic range. This application of deep learning provides an avenue towards the development of predictive control systems for powered limbs aimed at optimizing prosthetic ankle torque. </p>

show abstract

Predicting biological joint moment during multiple ambulation tasks

Cited by 26 publications

References 34 publications

Discovering individual-specific gait signatures from data-driven models of neuromechanical dynamics

Discovering individual-specific gait signatures from data-driven models of neuromechanical dynamics

Estimation of Lower Extremity Joint Moments and 3D Ground Reaction Forces Using IMU Sensors in Multiple Walking Conditions: A Deep Learning Approach

Using Deep Learning Models to Predict Prosthetic Ankle Torque

Contact Info

Product

Resources

About