A Machine Learning and Wearable Sensor Based Approach to Estimate External Knee Flexion and Adduction Moments During Various Locomotion Tasks

Stetter, Bernd; Krafft, Frieder C.; Ringhof, Steffen; Stein, Thorsten; Sell, Stefan

doi:10.3389/fbioe.2020.00009

Cited by 75 publications

(132 citation statements)

References 47 publications

Supporting

Mentioning

110

Contrasting

Order By: Relevance

“…Specifically, in knee OA, KAM remains a strong predictor of structural progression [ 10 , 11 , 12 ] and a key outcome examined for gait retraining [ 98 , 99 ] and surgical interventions [ 100 ]. Previous research has demonstrated the ability of inertial sensors to estimate joint moments, but this has most often occurred in healthy populations [ 101 , 102 , 103 ]. Alternatively, we identified three studies which examined KAM [ 41 , 83 , 86 ], as well as one examining joint contact forces [ 36 ] and one examining ankle joint moments [ 70 ].…”

Section: Discussionmentioning

confidence: 99%

Wearable Inertial Sensors for Gait Analysis in Adults with Osteoarthritis—A Scoping Review

Kobsar

Masood

Khan

et al. 2020

Sensors

View full text Add to dashboard Cite

Our objective was to conduct a scoping review which summarizes the growing body of literature using wearable inertial sensors for gait analysis in lower limb osteoarthritis. We searched six databases using predetermined search terms which highlighted the broad areas of inertial sensors, gait, and osteoarthritis. Two authors independently conducted title and abstract reviews, followed by two authors independently completing full-text screenings. Study quality was also assessed by two independent raters and data were extracted by one reviewer in areas such as study design, osteoarthritis sample, protocols, and inertial sensor outcomes. A total of 72 articles were included, which studied the gait of 2159 adults with osteoarthritis (OA) using inertial sensors. The most common location of OA studied was the knee (n = 46), followed by the hip (n = 22), and the ankle (n = 7). The back (n = 41) and the shank (n = 40) were the most common placements for inertial sensors. The three most prevalent biomechanical outcomes studied were: mean spatiotemporal parameters (n = 45), segment or joint angles (n = 33), and linear acceleration magnitudes (n = 22). Our findings demonstrate exceptional growth in this field in the last 5 years. Nevertheless, there remains a need for more longitudinal study designs, patient-specific models, free-living assessments, and a push for “Code Reuse” to maximize the unique capabilities of these devices and ultimately improve how we diagnose and treat this debilitating disease.

show abstract

Section: Discussionmentioning

confidence: 99%

Wearable Inertial Sensors for Gait Analysis in Adults with Osteoarthritis—A Scoping Review

Kobsar

Masood

Khan

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

“…A variety of methods have been proposed for this purpose, and reviewed by Ancillao et al [ 9 ]. To overcome accuracy limitations and the restricted subsets of parameters that can be determined, researchers have focused on applying machine learning methods to improve the prediction of GRFs, joint angles and joint moments [ 2 , 10 , 11 , 12 , 13 , 14 , 15 ], with initial efforts focused on predicting smaller subsets of data, such as single GRF and joint moment components [ 10 , 11 , 12 ], or in the case of Stetter et al [ 13 ] by predicting sagittal and frontal plane moments in isolation. Very recently gait researchers have trained machine learning models to predict all component joint angles [ 14 , 15 ] and moments across all lower limb joints [ 14 ].…”

Section: Introductionmentioning

confidence: 99%

“…They provide baseline information for the predictability of all types of data. IMU sensor data has been used as input into this type of ANN for the prediction of joint angles and joint moments [ 12 , 13 , 14 , 17 ].…”

Section: Introductionmentioning

confidence: 99%

A Comparison of Three Neural Network Approaches for Estimating Joint Angles and Moments from Inertial Measurement Units

Mundt

Johnson

Potthast

et al. 2021

Sensors

View full text Add to dashboard Cite

The application of artificial intelligence techniques to wearable sensor data may facilitate accurate analysis outside of controlled laboratory settings—the holy grail for gait clinicians and sports scientists looking to bridge the lab to field divide. Using these techniques, parameters that are difficult to directly measure in-the-wild, may be predicted using surrogate lower resolution inputs. One example is the prediction of joint kinematics and kinetics based on inputs from inertial measurement unit (IMU) sensors. Despite increased research, there is a paucity of information examining the most suitable artificial neural network (ANN) for predicting gait kinematics and kinetics from IMUs. This paper compares the performance of three commonly employed ANNs used to predict gait kinematics and kinetics: multilayer perceptron (MLP); long short-term memory (LSTM); and convolutional neural networks (CNN). Overall high correlations between ground truth and predicted kinematic and kinetic data were found across all investigated ANNs. However, the optimal ANN should be based on the prediction task and the intended use-case application. For the prediction of joint angles, CNNs appear favourable, however these ANNs do not show an advantage over an MLP network for the prediction of joint moments. If real-time joint angle and joint moment prediction is desirable an LSTM network should be utilised.

show abstract

“…For example, estimation of KFM from the proposed technique was comparable to a neural network (NN) using EMG inputs ( r 2 = 0.81 vs. 0.76 for IMC-GP) [71] and a linear model using data from an instrumented insole ( r = 0.89) [72]. In more recent developments, NN-based architectures with IMU inputs have been used to estimate KFM with 1.14 %BW•H RMSE and r = 0.98 in a four-sensor, four-segment configuration [73] and with 18.4 %range RMSE and r = 0.72 in a two-sensor, two-segment configuration [74]. In addition to characterization of KFM, IMC-GP provides complementary insight into the function and loading of individual muscles which are not modeled in machine learning techniques.…”

Section: Discussionmentioning

confidence: 99%

Wearables-Only Analysis of Muscle and Joint Mechanics: An EMG-Driven Approach

Gurchiek

Donahue

Fiorentino

et al. 2021

Preprint

View full text Add to dashboard Cite

Complex sensor arrays prohibit practical deployment of existing wearables-based algorithms for free-living analysis of muscle and joint mechanics. Machine learning techniques have been proposed as a potential solution, however, they are less interpretable and generalizable when compared to physics-based techniques. Herein, we propose a hybrid method utilizing inertial sensor- and electromyography (EMG)-driven simulation of muscle contraction to characterize knee joint and muscle mechanics during walking gait. Machine learning is used only to map a subset of measured muscle excitations to a full set thereby reducing the number of required sensors. We demonstrate the utility of the approach for estimating net knee flexion moment (KFM) as well as individual muscle moment and work during the stance phase of gait across nine unimpaired subjects. Across all subjects, KFM was estimated with 0.91 %BW·H RMSE and strong correlations (r = 0.87) compared to ground truth inverse dynamics analysis. Estimates of individual muscle moments were strongly correlated (r = 0.81-0.99) with a reference EMG-driven technique using optical motion capture and a full set of electrodes as were estimates of muscle work (r = 0.88-0.99). Implementation of the proposed technique in the current work included instrumenting only three muscles with surface electrodes (lateral and medial gastrocnemius and vastus medialis) and both the thigh and shank segments with inertial sensors. These sensor locations permit instrumentation of a knee brace/sleeve facilitating a practically deployable mechanism for monitoring muscle and joint mechanics with performance comparable to the current state-of-the-art.

show abstract

A Machine Learning and Wearable Sensor Based Approach to Estimate External Knee Flexion and Adduction Moments During Various Locomotion Tasks

Cited by 75 publications

References 47 publications

Wearable Inertial Sensors for Gait Analysis in Adults with Osteoarthritis—A Scoping Review

Wearable Inertial Sensors for Gait Analysis in Adults with Osteoarthritis—A Scoping Review

A Comparison of Three Neural Network Approaches for Estimating Joint Angles and Moments from Inertial Measurement Units

Wearables-Only Analysis of Muscle and Joint Mechanics: An EMG-Driven Approach

Contact Info

Product

Resources

About