Machine learning approach to predict center of pressure trajectories in a complete gait cycle: a feedforward neural network vs. LSTM network

Choi, Ahnryul; Jung, Hyunwoo; Lee, Ki Young; Lee, Sangsik; Mun, Joung Hwan

doi:10.1007/s11517-019-02056-0

Cited by 26 publications

(17 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As experimental equipment, six MCam2 cameras (VICON, Oxford Metrics, Oxford, UK) and 2 OR6-6-2000 ground reaction forces (AMTI Inc., Newton, MA, USA) were used. Each system was operated at sampling rates of 120 Hz and 1080 Hz, respectively [24,26,27]. Participants wore a Pedar-X insole system (Pedar Mobile, Novel Electronics Inc., GmbH, Munich, Germany).…”

Section: Subjects Apparatus and Lifting Experimentsmentioning

confidence: 99%

Machine Learning Model to Estimate Net Joint Moments during Lifting Task Using Wearable Sensors: A Preliminary Study for Design of Exoskeleton Control System

et al. 2021

Self Cite

View full text Add to dashboard Cite

Accurately measuring the lower extremities and L5/S1 moments is important since L5/S1 moments are the principal parameters that measure the risk of musculoskeletal diseases during lifting. In this study, protocol that predicts lower extremities and L5/S1 moments with an insole sensor was proposed to replace the prior methods that have spatial constraints. The protocol is hierarchically composed of a classification model and a regression model to predict joint moments. Additionally, a single LSTM model was developed to compare with proposed protocol. To optimize hyperparameters of the machine learning model and input feature, Bayesian optimization method was adopted. As a result, the proposed protocol showed a relative root mean square error (rRMSE) of 8.06~13.88% while the single LSTM showed 9.30~18.66% rRMSE. This protocol in this research is expected to be a starting point for developing a system for estimating the lower extremity and L5/S1 moment during lifting that can replace the complex prior method and adopted to workplace environments. This novel study has the potential to precisely design a feedback iterative control system of an exoskeleton for the appropriate generation of an actuator torque.

show abstract

Section: Subjects Apparatus and Lifting Experimentsmentioning

confidence: 99%

Machine Learning Model to Estimate Net Joint Moments during Lifting Task Using Wearable Sensors: A Preliminary Study for Design of Exoskeleton Control System

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…They found hybrid approaches incorporating domain knowledge by feature selection into the model to be helpful for accurate predictions. Different approaches were used to automatically derive the most relevant information from the data and reduce redundancy: an exhaustive greedy algorithm [ 16 ] has been used to find the most relevant features to predict the centre of pressure trajectory [ 17 ] or the 3D ground reaction force using pressure insoles [ 18 ]. The correlation between different inputs has been determined to exclude highly correlated input features from 3D segment angles for the prediction of the knee adduction moment [ 19 ].…”

Section: Introductionmentioning

confidence: 99%

Artificial Neural Networks in Motion Analysis—Applications of Unsupervised and Heuristic Feature Selection Techniques

Mundt

Koeppe

Bamer

et al. 2020

Sensors

View full text Add to dashboard Cite

The use of machine learning to estimate joint angles from inertial sensors is a promising approach to in-field motion analysis. In this context, the simplification of the measurements by using a small number of sensors is of great interest. Neural networks have the opportunity to estimate joint angles from a sparse dataset, which enables the reduction of sensors necessary for the determination of all three-dimensional lower limb joint angles. Additionally, the dimensions of the problem can be simplified using principal component analysis. Training a long short-term memory neural network on the prediction of 3D lower limb joint angles based on inertial data showed that three sensors placed on the pelvis and both shanks are sufficient. The application of principal component analysis to the data of five sensors did not reveal improved results. The use of longer motion sequences compared to time-normalised gait cycles seems to be advantageous for the prediction accuracy, which bridges the gap to real-time applications of long short-term memory neural networks in the future.

show abstract

“…However, the input matrix had a size of 38,332 (number of frames: 38,332 = 5 subjects × 74 steps × 148 frames × 70% for training dataset) × 6-24 (number of input features, which depended on the IMU combination). The input matrix size might be considered comparable to the previous study of estimating the inclination angle between the center of mass and COP (36,000 × 9), which involved 24 subjects (Choi et al, 2019a). However, the input matrix provided sufficient within-subject variability, but not betweensubject variability, due to the limited number of subjects.…”

Section: Discussionmentioning

confidence: 99%

“…Leporace et al (2015) used multilayer perceptron networks to predict tri-axial ground reaction force (GRF) based on IMU data. Choi et al (2019a) compared the performance of the feed-forward artificial neural network (FFANN) and long short-term memory (LSTM) in the prediction of complete gait cycle COP based on single-stance COP. Of the various machine learning algorithms, LSTM is often used in time series data and showed better performance than others (Gers and Schmidhuber, 2000;Zhao et al, 2018).…”

Section: Introductionmentioning

confidence: 99%

Multiple Inertial Measurement Unit Combination and Location for Center of Pressure Prediction in Gait

Chen

Hsu

et al. 2020

Front. Bioeng. Biotechnol.

View full text Add to dashboard Cite

Center of pressure (COP) during a gait cycle indicates crucial information with regard to fall risk such as balance capacity. The drawbacks of conventional research instruments include inconvenient use during activities of daily living and expensive costs. The present study illustrates the promising fall-relevant information predicted by acceleration and angular velocity data from different placement sensors with machine learning techniques. This approach is inspired by the emerging machine learning technique, specifically the long short-term memory (LSTM), which is often used in time series data and aims to decrease the burden of the user while using the novel wearable technology. The Jaccard similarity coefficient, which implies the consistency of profile alignment between prediction and real situation, achieved 94% accuracy in the walking direction. Furthermore, the number of sensors used and the placement influenced the feasibility of an application. The outcome revealed that the accuracy could exceed 90% with only one sensor placed on the foot in the walking direction, and the toe would be the best location for sensor placement. To examine the performance of machine learning, the current study employed two parameters from different perspectives. One is a commonly used parameter, which represented the error, and the other investigated the similarity between the prediction and ground truth. From a similarity perspective, the parameter can be used as a metric to assess the consistency of profile alignment.

show abstract

Machine learning approach to predict center of pressure trajectories in a complete gait cycle: a feedforward neural network vs. LSTM network

Cited by 26 publications

References 40 publications

Machine Learning Model to Estimate Net Joint Moments during Lifting Task Using Wearable Sensors: A Preliminary Study for Design of Exoskeleton Control System

Machine Learning Model to Estimate Net Joint Moments during Lifting Task Using Wearable Sensors: A Preliminary Study for Design of Exoskeleton Control System

Artificial Neural Networks in Motion Analysis—Applications of Unsupervised and Heuristic Feature Selection Techniques

Multiple Inertial Measurement Unit Combination and Location for Center of Pressure Prediction in Gait

Contact Info

Product

Resources

About