Machine learning for rapid estimation of lower extremity muscle and joint loading during activities of daily living

Knee-Loading Predictions with Neural Networks Improve Finite Element Modeling Classifications of Knee Osteoarthritis: Data from the Osteoarthritis Initiative

Physics-based modeling methods have the potential to investigate the mechanical factors associated with knee osteoarthritis (OA) and predict the future radiographic condition of the joint. However, it remains unclear what level of detail is optimal in these methods to achieve accurate prediction results in cohort studies. In this work, we extended a template-based finite element (FE) method to include the lateral and medial compartments of the tibiofemoral joint and simulated the mechanical responses of 97 knees under three conditions of gait loading. Furthermore, the effects of variations in cartilage thickness and failure equation on predicted cartilage degeneration were investigated. Our results showed that using neural network-based estimations of peak knee loading provided classification performances of 0.70 (AUC, p < 0.05) in distinguishing between knees that developed severe OA or mild OA and knees that did not develop OA eight years after a healthy radiographic baseline. However, FE models incorporating subject-specific femoral and tibial cartilage thickness did not improve this classification performance, suggesting there exists an optimal point between personalized loading and geometry for discrimination purposes. In summary, we proposed a modeling framework that streamlines the rapid generation of individualized knee models achieving promising classification performance while avoiding motion capture and cartilage image segmentation.

Integrating musculoskeletal simulation and machine learning: a hybrid approach for personalized ankle-foot exoskeleton assistance strategies

Introduction: Lower limb exoskeletons have shown considerable potential in assisting human walking, particularly by reducing metabolic cost (MC), leading to a surge of interest in this field in recent years. However, owing to significant individual differences and the uncertainty of movements, challenges still exist in the personalized design and control of exoskeletons in human-robot interactions. Methods: In this study, we propose a hybrid data-driven approach that integrates musculoskeletal simulation with machine learning technology to customize personalized assistance strategies efficiently and adaptively for ankle-foot exoskeletons. First, optimal assistance strategies that can theoretically minimize MC, were derived from forward muscle-driven simulations on an open-source dataset. Then, a neural network was utilized to explore the relationships among different individuals, movements, and optimal strategies, thus developing a predictive model. Results: With respect to transfer learning, our approach exhibited effectiveness and adaptability when faced with new individuals and movements. The simulation results further indicated that our approach successfully reduced the MC of calf muscles by approximately 20% compared to normal walking conditions. Discussion: This hybrid approach offers an alternative for personalizing assistance strategy that may further guide exoskeleton design.

Predicting Knee Joint Contact Forces During Normal Walking Using Kinematic Inputs With a Long-Short Term Neural Network

Knee joint contact forces are commonly estimated via surrogate measures (i.e., external knee adduction moments or musculoskeletal modeling). Despite its capabilities, modeling is not optimal for clinicians or persons with limited experience. The purpose of this study was to design a novel prediction method for knee joint contact forces that is simplistic in terms of required inputs. This study included marker trajectories and instrumented knee forces during normal walking from the "Grand Challenge" (n = 6) and "CAMS" (n = 2) datasets. Inverse kinematics were used to derive stance phase hip (sagittal, frontal, transverse), knee (sagittal, frontal), ankle (sagittal), and trunk (frontal) kinematics. A long-short term memory network (LSTM) was created using matlab to predict medial and lateral knee force waveforms using combinations of the kinematics. The Grand Challenge and CAMS datasets trained and tested the network, respectively. Musculoskeletal modeling forces were derived using static optimization and joint reaction tools in OpenSim. Waveform accuracy was determined as the proportion of variance and root-mean-square error between network predictions and in vivo data. The LSTM network was highly accurate for medial forces (R2 = 0.77, RMSE = 0.27 BW) and required only frontal hip and knee and sagittal hip and ankle kinematics. Modeled medial force predictions were excellent (R2 = 0.77, RMSE = 0.33 BW). Lateral force predictions were poor for both methods (LSTM R2 = 0.18, RMSE = 0.08 BW; modeling R2 = 0.21, RMSE = 0.54 BW). The designed LSTM network outperformed most reports of musculoskeletal modeling, including those reached in this study, revealing knee joint forces can accurately be predicted by using only kinematic input variables.

Machine Learning Approaches for 3D Motion Synthesis and Musculoskeletal Dynamics Estimation: A Survey

The inference of 3D motion and dynamics of the human musculoskeletal system has traditionally been solved using physics-based methods that exploit physical parameters to provide realistic simulations. Yet, such methods suffer from computational complexity and reduced stability, hindering their use in computer graphics applications that require real-time performance. With the recent explosion of data capture (mocap, video) machine learning (ML) has started to become popular as it is able to create surrogate models harnessing the huge amount of data stemming from various sources, minimizing computational time (instead of resource usage), and most importantly, approximate real-time solutions. The main purpose of this paper is to provide a review and classification of the most recent works regarding motion prediction, motion synthesis as well as musculoskeletal dynamics estimation problems using ML techniques, in order to offer sufficient insight into the state-of-the-art and draw new research directions. While the study of motion may appear distinct to musculoskeletal dynamics, these application domains provide jointly the link for more natural computer graphics character animation, since ML-based musculoskeletal dynamics estimation enables modeling of more long-term, temporally evolving, ergonomic effects, while offering automated and fast solutions. Overall, our review offers an in-depth presentation and classification of ML applications in human motion analysis, unlike previous survey articles focusing on specific aspects of motion prediction.

Prediction of Achilles Tendon Force During Common Motor Tasks From Markerless Video

Remodeling of the Achilles tendon (AT) is partly driven by its mechanical environment. AT force can be estimated with neuromusculoskeletal (NMSK) modeling; however, the complex experimental setup required to perform the analyses confines use to the laboratory. We developed task-specific long short-term memory (LSTM) neural networks that employ markerless video data to predict the AT force during walking, running, countermovement jump, single-leg landing, and single-leg heel rise. The task-specific LSTM models were trained on pose estimation keypoints and corresponding AT force data from 16 subjects, calculated via an established NMSK modeling pipeline, and cross-validated using a leave-one-subject-out approach. As proof-of-concept, new motion data of one participant was collected with two smartphones and used to predict AT forces. The task-specific LSTM models predicted the time-series AT force using synthesized pose estimation data with root mean square error (RMSE) ≤ 526 N, normalized RMSE (nRMSE) ≤ 0.21 , R 2 ≥ 0.81 . Walking task resulted the most accurate with RMSE = 189±62 N; nRMSE = 0.11±0.03 , R 2 = 0.92±0.04 . AT force predicted with smartphones video data was physiologically plausible, agreeing in timing and magnitude with established force profiles. This study demonstrated the feasibility of using low-cost solutions to deploy complex biomechanical analyses outside the laboratory.

Prediction of medial knee contact force using multisource fusion recurrent neural network and transfer learning

Estimation of knee contact force (KCF) during gait provides essential information to evaluate knee joint function. Machine learning has been employed to estimate KCF because of the advantages of low computational cost and real-time. However, the existing machine learning models do not adequately consider gait-related data's temporal-dependent, multidimensional, and highly heterogeneous nature. This study is aimed at developing a multisource fusion recurrent neural network to predict the medial condyle KCF. First, a multisource fusion long short-term memory (MF-LSTM) model was established. Then, we developed a transfer learning strategy based on the MF-LSTM model for subject-specific medial KCF prediction. Four subjects with instrumented tibial prostheses were obtained from the literature. The results showed that the MF-LSTM model could predict medial KCF to a certain high level of accuracy (the mean of ρ = 0.970). The transfer learning model improved the prediction accuracy (the mean of ρ = 0.987). This study shows that the MF-LSTM model is a powerful and accurate computational tool for medial KCF prediction. Introducing transfer learning techniques could further improve the prediction performance for the target subject. This coupling strategy can help clinicians accurately estimate and track joint contact forces in real time.

Differences in running technique between runners with better and poorer running economy and lower and higher milage: An artificial neural network approach

BACKGROUND

Prior studies investigated selected discrete sagittal-plane outcomes (e.g., peak knee flexion) in relation to running economy, hereby discarding the potential relevance of running technique parameters during noninvestigated phases of the gait cycle and in other movement planes.

PURPOSE

Investigate which components of running technique distinguish groups of runners with better and poorer economy and higher and lower weekly running distance using an artificial neural network (ANN) approach with layer-wise relevance propagation.

METHODS

Forty-one participants (22 males and 19 females) ran at 2.78 m∙s-1 while three-dimensional kinematics and gas exchange data were collected. Two groups were created that differed in running economy or weekly training distance. The three-dimensional kinematic data were used as input to an ANN to predict group allocations. Layer-wise relevance propagation was used to determine the relevance of three-dimensional kinematics for group classification.

RESULTS

The ANN classified runners in the correct economy or distance group with accuracies of up to 62% and 71%, respectively. Knee, hip, and ankle flexion were most relevant to both classifications. Runners with poorer running economy showed higher knee flexion during swing, more hip flexion during early stance, and more ankle extension after toe-off. Runners with higher running distance showed less trunk rotation during swing.

CONCLUSION

The ANN accuracy was moderate when predicting whether runners had better, or poorer running economy, or had a higher or lower weekly training distance based on their running technique. The kinematic components that contributed the most to the classification may nevertheless inform future research and training.

Integrating an LSTM framework for predicting ankle joint biomechanics during gait using inertial sensors

The ankle joint plays a crucial role in gait, facilitating the articulation of the lower limb, maintaining foot-ground contact, balancing the body, and transmitting the center of gravity. This study aimed to implement long short-term memory (LSTM) networks for predicting ankle joint angles, torques, and contact forces using inertial measurement unit (IMU) sensors. Twenty-five healthy participants were recruited. Two IMU sensors were attached to the foot dorsum and the vertical axis of the distal anteromedial tibia in the right lower limb to record acceleration and angular velocity during running. We proposed a LSTM-MLP (multilayer perceptron) model for training time-series data from IMU sensors and predicting ankle joint biomechanics. The model underwent validation and testing using a custom nested k-fold cross-validation process. The average values of the coefficient of determination (R2), mean absolute error (MAE), and mean squared error (MSE) for ankle dorsiflexion joint and moment, subtalar inversion joint and moment, and ankle joint contact forces were 0.89 ± 0.04, 0.75 ± 1.04, and 2.96 ± 4.96 for walking, and 0.87 ± 0.07, 0.88 ± 1.26, and 4.1 ± 7.17 for running, respectively. This study demonstrates that IMU sensors, combined with LSTM neural networks, are invaluable tools for evaluating ankle joint biomechanics in lower limb pathological diagnosis and rehabilitation, offering a cost-effective and versatile alternative to traditional experimental settings.

Sagittal plane knee kinematics can be measured during activities of daily living following total knee arthroplasty with two IMU

Knee function is rarely measured objectively during functional tasks following total knee arthroplasty. Inertial measurement units (IMU) can measure knee kinematics and range of motion (ROM) during dynamic activities and offer an easy-to-use system for knee function assessment post total knee arthroplasty. However, IMU must be validated against gold standard three-dimensional optical motion capture systems (OMC) across a range of tasks if they are to see widespread uptake. We computed knee rotations and ROM from commercial IMU sensor measurements during walking, squatting, sit-to-stand, stair ascent, and stair descent in 21 patients one-year post total knee arthroplasty using two methods: direct computation using segment orientations (r_IMU), and an IMU-driven iCloud-based interactive lower limb model (m_IMU). This cross-sectional study compared computed knee angles and ROM to a gold-standard OMC and inverse kinematics method using Pearson's correlation coefficient (R) and root-mean-square-differences (RMSD). The r_IMU and m_IMU methods estimated sagittal plane knee angles with excellent correlation (>0.95) compared to OMC for walking, squatting, sit-to-stand, and stair-ascent, and very good correlation (>0.90) for stair descent. For squatting, sit-to-stand, and walking, the mean RMSD for r_IMU and m_IMU compared to OMC were <4 degrees, < 5 degrees, and <6 degrees, respectively but higher for stair ascent and descent (~12 degrees). Frontal and transverse plane knee kinematics estimated using r_IMU and m_IMU showed poor to moderate correlation compared to OMC. There were no differences in ROM measurements during squatting, sit-to-stand, and walking across the two methods. Thus, IMUs can measure sagittal plane knee angles and ROM with high accuracy for a variety of tasks and may be a useful in-clinic tool for objective assessment of knee function following total knee arthroplasty.

Machine learning-based prediction of hip joint moment in healthy subjects, patients and post-operative subjects

The application of machine learning in the field of motion capture research is growing rapidly. The purpose of the study is to implement a long-short term memory (LSTM) model able to predict sagittal plane hip joint moment (HJM) across three distinct cohorts (healthy controls, patients and post-operative patients) starting from 3D motion capture and force data. Statistical parametric mapping with paired samples t-test was performed to compare machine learning and inverse dynamics HJM predicted values, with the latter used as gold standard. The results demonstrated favorable model performance on each of the three cohorts, showcasing its ability to successfully generalize predictions across diverse cohorts.

On the prediction of tibiofemoral contact forces for healthy individuals and osteoarthritis patients during gait: a comparative study of regression methods

Knee osteoarthritis (OA) is a public health problem affecting millions of people worldwide. The intensity of the tibiofemoral contact forces is related to cartilage degeneration, and so is the importance of quantifying joint loads during daily activities. Although simulation with musculoskeletal models has been used to calculate joint loads, it demands high-cost equipment and a very time-consuming process. This study aimed to evaluate consolidated machine learning algorithms to predict tibiofemoral forces during gait analysis of healthy individuals and knee OA patients. Also, we evaluated three different datasets to train each model, considering different combinations of primary kinematic and kinetic data, and post-processing data. We evaluated 14 patients with severe unilateral knee OA and 14 healthy individuals during 3-5 gait trials. Data were split into 70% and 30% of the samples as training and test data. Test data was independently evaluated considering a mixture of pathological and healthy individuals, and only OA and Control patients. The main results showed that accurate predictions of the tibiofemoral contact forces were achieved using machine learning methods and that the predictions were sensitive to changes in the input data as training. The present study provided insights into the most promising regressions methods to predict knee contact forces representing an important starting point for the broader application of biomechanical analysis in clinical environments.

Prediction of knee biomechanics with different tibial component malrotations after total knee arthroplasty: conventional machine learning vs. deep learning

The precise alignment of tibiofemoral components in total knee arthroplasty is a crucial factor in enhancing the longevity and functionality of the knee. However, it is a substantial challenge to quickly predict the biomechanical response to malrotation of tibiofemoral components after total knee arthroplasty using musculoskeletal multibody dynamics models. The objective of the present study was to conduct a comparative analysis between a deep learning method and four conventional machine learning methods for predicting knee biomechanics with different tibial component malrotation during a walking gait after total knee arthroplasty. First, the knee contact forces and kinematics with different tibial component malrotation in the range of ±5° in the three directions of anterior/posterior slope, internal/external rotation, and varus/valgus rotation during a walking gait after total knee arthroplasty were calculated based on the developed musculoskeletal multibody dynamics model. Subsequently, deep learning and four conventional machine learning methods were developed using the above 343 sets of biomechanical data as the dataset. Finally, the results predicted by the deep learning method were compared to the results predicted by four conventional machine learning methods. The findings indicated that the deep learning method was more accurate than four conventional machine learning methods in predicting knee contact forces and kinematics with different tibial component malrotation during a walking gait after total knee arthroplasty. The deep learning method developed in this study enabled quickly determine the biomechanical response with different tibial component malrotation during a walking gait after total knee arthroplasty. The proposed method offered surgeons and surgical robots the ability to establish a calibration safety zone, which was essential for achieving precise alignment in both preoperative surgical planning and intraoperative robotic-assisted surgical navigation.

Glenohumeral joint force prediction with deep learning

Deep learning models (DLM) are efficient replacements for computationally intensive optimization techniques. Musculoskeletal models (MSM) typically involve resource-intensive optimization processes for determining joint and muscle forces. Consequently, DLM could predict MSM results and reduce computational costs. Within the total shoulder arthroplasty (TSA) domain, the glenohumeral joint force represents a critical MSM outcome as it can influence joint function, joint stability, and implant durability. Here, we aimed to employ deep learning techniques to predict both the magnitude and direction of the glenohumeral joint force. To achieve this, 959 virtual subjects were generated using the Markov-Chain Monte-Carlo method, providing patient-specific parameters from an existing clinical registry. A DLM was constructed to predict the glenohumeral joint force components within the scapula coordinate system for the generated subjects with a coefficient of determination of 0.97, 0.98, and 0.98 for the three components of the glenohumeral joint force. The corresponding mean absolute errors were 11.1, 12.2, and 15.0 N, which were about 2% of the maximum glenohumeral joint force. In conclusion, DLM maintains a comparable level of reliability in glenohumeral joint force estimation with MSM, while drastically reducing the computational costs.

Estimation of lower limb joint moments based on the inverse dynamics approach: a comparison of machine learning algorithms for rapid estimation

The aim of this study is to estimate the joint moments of the ankle, knee, and hip joints during walking. A sit-to-stand (STS) movement analysis was first performed on 20 participants with different anthropometric characteristics. Then, analysis of the dynamics of the STS motion was used to develop a biomechanical model. Decision tree (DT), linear regression (LR), support vector machine (SVM), random forest (RF), and three deep learning (DL) algorithms and deep neural network (DNN), long-short-term memory (LSTM), and convolutional neural network (CNN) are examined in this work to estimate three joint moments: ankle, knee, and hip. The results of the seven algorithms were evaluated using four statistical benchmarks: MSR, RMSE, correlation coefficient (R), and MAE to find the most accurate one. The results show that the most successful algorithms were LSTM in estimating knee, hip, and ankle joint moments using 19 and 7 inputs. The R value was 0.9990 using 19 inputs and 0.9972 using 7 inputs. The other algorithms have a correlation coefficient (R) success of 0.9902, 0.9770, 0.9884, 0.9577, 0.9786, and 0.9022 for RF, CNN, DT, DNN, SVM, and LR, respectively. The prediction of joint moments plays a crucial role in the design of the biomechanical system with the desired mechanical properties. Especially, the need has arisen to predict joint moments in a shorter time to utilize in real-time active prosthesis/orthosis controllers.

Prediction of Knee Joint Compartmental Loading Maxima Utilizing Simple Subject Characteristics and Neural Networks

Joint loading may affect the development of osteoarthritis, but patient-specific load estimation requires cumbersome motion laboratory equipment. This reliance could be eliminated using artificial neural networks (ANNs) to predict loading from simple input predictors. We used subject-specific musculoskeletal simulations to estimate knee joint contact forces for 290 subjects during over 5000 stance phases of walking and then extracted compartmental and total joint loading maxima from the first and second peaks of the stance phase. We then trained ANN models to predict the loading maxima from predictors that can be measured without motion laboratory equipment (subject mass, height, age, gender, knee abduction-adduction angle, and walking speed). When compared to the target data, our trained models had NRMSEs (RMSEs normalized to the mean of the response variable) between 0.14 and 0.42 and Pearson correlation coefficients between 0.42 and 0.84. The loading maxima were predicted most accurately using the models trained with all predictors. We demonstrated that prediction of knee joint loading maxima may be possible without laboratory-measured motion capture data. This is a promising step in facilitating knee joint loading predictions in simple environments, such as a physician's appointment. In future, the rapid measurement and analysis setup could be utilized to guide patients in rehabilitation to slow development of joint disorders, such as osteoarthritis.

Accuracy of Patient-Specific Instrument for Cylindrical Axis Implementation in Kinematically Aligned Total Knee Arthroplasty

Background

In kinematically aligned total knee arthroplasty (KA-TKA), the cylindrical axis (CA) is very important in restoring the native joint line and kinematics of the pre-arthritic knee. This study aimed to determine the accuracy of patient-specific instrument (PSI) for restoring the CA for femoral bone resection in KA-TKA.

Methods

Thirty KA-TKAs were performed using a computed tomography (CT)-based PSI system. Data from preoperative CT were reconstructed into three-dimensional (3D) models using 3D-planning software. The CA was created by connecting the centers of each virtual sphere to the medial and lateral femoral condyles using computer software. Femoral bone resection of the distal and posterior condyles was performed parallel to the sagittal planes of the CA. The thickness of the CA-referenced bone resection was determined based on the thickness necessary for the respective regions of the femoral component. The PSI was manufactured to locate the guide pin for a conventional cutting block. The accuracy of PSI for KA-TKA was evaluated as the absolute error between the preoperatively predicted thickness and the intraoperative measurements in each of the four regions, as well as the difference in error between distal-medial (DM) and posterior-medial (PM) and between distal-lateral (DL) and posterior-lateral (PL).

Results

The differences in thickness of bone cut in the DM, DL, PM, and PL were 0.79 ± 0.39 mm (range, -1.20 to 1.50), 0.70 ± 0.42 mm (range, -1.50 to 1.50), 0.80 ± 0.46 mm (range, -0.80 to 1.50), and 0.75 ± 0.47 mm (range, -2.10 to 1.40), respectively. There was no significant difference in the thickness error between DM and PM (p = 0.959) and between DL and PL (p = 0.812).

Conclusions

In KA-TKA, PSI was effective for accurate femoral bone resection based on virtually planned thickness.

Inverse distance weighting to rapidly generate large simulation datasets

Obtaining large biomechanical datasets for machine learning is an ongoing challenge. Physics-based simulations offer one approach for generating large datasets, but many simulation methods, such as computed muscle control (CMC), are computationally costly. In contrast, interpolation methods, such as inverse distance weighting (IDW), are computationally fast. We examined whether IDW is a low-cost and accurate approach for interpolating muscle activations from CMC.IDW was evaluated using lateral pinch simulations in OpenSim. Simulated pinch data were organized into grids of varying sparsity (high, medium, and low density), where each grid point represented the muscle activations associated with a unique combination of mass and height of a young adult. For each grid, muscle activations were calculated via CMC and IDW for 108 random mass-height pairs that were not coincident with simulation grid vertices. We evaluated the interpolation errors from IDW for each grid, as well as the sensitivity of lateral pinch force to these errors. The root mean square error (RMSE) associated with interpolated muscle activations decreased with increasing grid density and never exceeded 4%. While CMC received a target thumb-tip force of 40 N, errors from the interpolated muscle activations never impacted the simulated force magnitude by more than 0.1 N. Furthermore, the computation time for CMC simulations averaged 4.22 core-minutes, while IDW averaged 0.95 core-seconds per mass-height pair.These results indicate IDW is a practical approach for rapidly estimating muscle activations from sparse CMC datasets. Future works could adapt our IDW approach to evaluate other tasks, biomechanical features, and/or populations.

Lower body kinematics estimation during walking using an accelerometer

Measuring and predicting accurate joint angles are important to developing analytical tools to gauge users' progress. Such measurement is usually performed in laboratory settings, which is difficult and expensive. So, the aim of this study was continuous estimation of lower limb joint angles during walking using an accelerometer and random forest (RF). Thus, 73 subjects (26 women and 47 men) voluntarily participated in this study. The subjects walked at the slow, moderate, and fast speeds on a walkway, which was covered with 10 Vicon camera. Acceleration was used as input for a RF to estimate ankle, knee, and hip angles (in transverse, frontal, and sagittal planes). Pearson correlation coefficient (r) and Mean Square Error (MSE) were computed between the experimental and estimated data. Paired statistical parametric mapping (SPM) t-test was used to compare the experimental and estimated data throughout gait cycle. The results of this study showed that the MSE of joint angles between the experimental and estimated data ranged from 0.04 to 24.29 and r > 0.91. Moreover, the findings of SPM indicated that there was no significant difference between the experimental and estimated data of ankle, knee, and hip angles in all three planes throughout gait cycle. The results of our research developed a more accessible, portable procedure to quantifying lower limb joint angles by an accelerometer and RF. So, such wearable-based joint angles have the potential to be used in outside-laboratory settings to measure walking kinematics.

Estimation of Lower Extremity Muscle Activity in Gait Using the Wearable Inertial Measurement Units and Neural Network

The inertial measurement unit (IMU) has become more prevalent in gait analysis. However, it can only measure the kinematics of the body segment it is attached to. Muscle behaviour is an important part of gait analysis and provides a more comprehensive overview of gait quality. Muscle behaviour can be estimated using musculoskeletal modelling or measured using an electromyogram (EMG). However, both methods can be tasking and resource intensive. A combination of IMU and neural networks (NN) has the potential to overcome this limitation. Therefore, this study proposes using NN and IMU data to estimate nine lower extremity muscle activities. Two NN were developed and investigated, namely feedforward neural network (FNN) and long short-term memory neural network (LSTM). The results show that, although both networks were able to predict muscle activities well, LSTM outperformed the conventional FNN. This study confirms the feasibility of estimating muscle activity using IMU data and NN. It also indicates the possibility of this method enabling the gait analysis to be performed outside the laboratory environment with a limited number of devices.

Increased Femoral Anteversion Does Not Lead to Increased Joint Forces During Gait in a Cohort of Adolescent Patients

Orthopedic complications were previously reported for patients with increased femoral anteversion. A more comprehensive analysis of the influence of increased femoral anteversion on joint loading in these patients is required to better understand the pathology and its clinical management. Therefore, the aim was to investigate lower-limb kinematics, joint moments and forces during gait in adolescent patients with increased, isolated femoral anteversion compared to typically developing controls. Secondly, relationships between the joint loads experienced by the patients and different morphological and kinematic features were investigated. Patients with increased femoral anteversion (n = 42, 12.8 ± 1.9 years, femoral anteversion: 39.6 ± 6.9°) were compared to typically developing controls (n = 9, 12.0 ± 3.0 years, femoral anteversion: 18.7 ± 4.1°). Hip and knee joint kinematics and kinetics were calculated using subject-specific musculoskeletal models. Differences between patients and controls in the investigated outcome variables (joint kinematics, moments, and forces) were evaluated through statistical parametric mapping with Hotelling T2 and t-tests (α = 0.05). Canonical correlation analyses (CCAs) and regression analyses were used to evaluate within the patients’ cohort the effect of different morphological and kinematic predictors on the outcome variables. Predicted compressive proximo-distal loads in both hip and knee joints were significantly reduced in patients compared to controls. A gait pattern characterized by increased knee flexion during terminal stance (KneeFlex_tSt) was significantly correlated with hip and knee forces, as well as with the resultant force exerted by the quadriceps on the patella. On the other hand, hip internal rotation and in-toeing, did not affect the loads in the joints. Based on the finding of the CCAs and linear regression analyses, patients were further divided into two subgroups based KneeFlex_tSt. Patients with excessive KneeFlex_tSt presented a significantly higher femoral anteversion than those with normal KneeFlex_tSt. Patients with excessive KneeFlex_tSt presented significantly larger quadriceps forces on the patella and a larger posteriorly-oriented shear force at the knee, compared to patients with normal KneeFlex_tSt, but both patients’ subgroups presented only limited differences in terms of joint loading compared to controls. This study showed that an altered femoral morphology does not necessarily lead to an increased risk of joint overloading, but instead patient-specific kinematics should be considered.

Multi-Output Sequential Deep Learning Model for Athlete Force Prediction on a Treadmill Using 3D Markers

Reliable and innovative methods for estimating forces are critical aspects of biomechanical sports research. Using them, athletes can improve their performance and technique and reduce the possibility of fractures and other injuries. For this purpose, throughout this project, we proceeded to research the use of video in biomechanics. To refine this method, we propose an RNN trained on a biomechanical dataset of regular runners that measures both kinematics and kinetics. The model will allow analyzing, extracting, and drawing conclusions about continuous variable predictions through the body. It marks different anatomical and reflective points (96 in total, 32 per dimension) that will allow the prediction of forces (N) in three dimensions (Fx, Fy, Fz), measured on a treadmill with a force plate at different velocities (2.5 m/s, 3.5 m/s, 4.5 m/s). In order to obtain the best model, a grid search of different parameters that combined various types of layers (Simple, GRU, LSTM), loss functions (MAE, MSE, MSLE), and sampling techniques (down-sampling, up-sampling) helped obtain the best performing model (LSTM, MSE, down-sampling) achieved an average coefficient of determination of 0.68, although when excluding Fz it reached 0.92.

The Use of Synthetic IMU Signals in the Training of Deep Learning Models Significantly Improves the Accuracy of Joint Kinematic Predictions

Gait analysis based on inertial sensors has become an effective method of quantifying movement mechanics, such as joint kinematics and kinetics. Machine learning techniques are used to reliably predict joint mechanics directly from streams of IMU signals for various activities. These data-driven models require comprehensive and representative training datasets to be generalizable across the movement variability seen in the population at large. Bottlenecks in model development frequently occur due to the lack of sufficient training data and the significant time and resources necessary to acquire these datasets. Reliable methods to generate synthetic biomechanical training data could streamline model development and potentially improve model performance. In this study, we developed a methodology to generate synthetic kinematics and the associated predicted IMU signals using open source musculoskeletal modeling software. These synthetic data were used to train neural networks to predict three degree-of-freedom joint rotations at the hip and knee during gait either in lieu of or along with previously measured experimental gait data. The accuracy of the models’ kinematic predictions was assessed using experimentally measured IMU signals and gait kinematics. Models trained using the synthetic data out-performed models using only the experimental data in five of the six rotational degrees of freedom at the hip and knee. On average, root mean square errors in joint angle predictions were improved by 38% at the hip (synthetic data RMSE: 2.3°, measured data RMSE: 4.5°) and 11% at the knee (synthetic data RMSE: 2.9°, measured data RMSE: 3.3°), when models trained solely on synthetic data were compared to measured data. When models were trained on both measured and synthetic data, root mean square errors were reduced by 54% at the hip (measured + synthetic data RMSE: 1.9°) and 45% at the knee (measured + synthetic data RMSE: 1.7°), compared to measured data alone. These findings enable future model development for different activities of clinical significance without the burden of generating large quantities of gait lab data for model training, streamlining model development, and ultimately improving model performance.