Which method should we use to determine the hip joint center location in individuals with a high amount of soft tissue?

BACKGROUND: This study investigated the most accurate method for estimating the hip joint center position in clinical 3D gait analysis for young individuals with high amounts of soft tissue. We compared position estimates of five regression-based and two functional methods to the hip joint center position obtained through 3D free-hand ultrasound. METHODS: For this purpose, the data of 14 overweight or obese individuals with a mean age of 13.6 (SD 2.1 yrs) and a BMI of 36.5 (SD 7.1 kg/m2, range 26-52 kg/m2) who underwent standard clinical 3D gait analysis were used. The data of each participant were processed with five regression-based and two functional methods and compared to the hip joint center identified via 3D free-hand ultrasound. FINDINGS: The absolute location errors to 3D free-hand ultrasound for each anatomical plane and the Euclidean distances served as outcomes next to their effects on gait variables. The data suggest that regression-based methods are preferable to functional methods in this population, as the latter demonstrated the highest variability in accuracy with large errors for some individuals. INTERPRETATION: Based on our findings we recommend using the regression method presented by Hara et al. due to its superior overall accuracy of <9 mm on average in all planes and the lowest impact on kinematic and kinetic output variables. We do not recommend using the Harrington equations (single and multiple) in populations with high amounts of soft tissue as they require pelvic depth as input, which can be massively biased when a lot of soft tissue is present around the pelvis.

Normative 3D gait data of healthy adults walking at three different speeds on an instrumented treadmill in virtual reality.

A normative gait dataset of 246 healthy adults (122 men / 124 women, range in age 18-91 years, body weight 46.80-116.10 kg, height 1.53-1.97 m and BMI 18.25-35.63 kg/m2) is presented and publicly shared for three walking speed conditions. Raw and processed data are presented for each subject separately and for each walking speed, including data of every single step of both legs. The subject demographics and results from the physical examination are also presented which allows researchers and clinicians to create a self-selected reference group based on specific demographics. Besides the data per individual, data are also presented in age and gender groups. This provides a quick overview of healthy gait parameters which is relevant for use in clinical practice. Three dimensional gait analysis was performed at the Computer Assisted Rehabilitation Environment (CAREN) at the Maastricht University Medical Centre (MUMC+). Subjects walked on the instrumented treadmill surrounded with twelve 3D cameras, three 2D cameras and a virtual industrial environment projected on a 180° screen using the Human Body Lower Limb Model with trunk markers (HBM-II) as biomechanical model [1], [2]. Subjects walked at comfortable walking speed, 30% slower and 30% faster. These walking speed conditions were applied in a random sequence. Comfortable walking speed was determined using a RAMP protocol: subjects started to walk at 0.5m/s and every second the speed was increased with 0.01 m/s until the preferred speed was reached. The average of three repetitions was considered the comfortable speed. For each walking speed condition, 250 steps were recorded. The 3D gait data was collected using the D-flow CAREN software. For each subject, raw data of each walking speed condition is provided in .mox files, including the output from the model such as subject data (e.g. gender, body mass, knee and ankle width), center of mass (CoM), marker and force data, kinematic data (joint angles) and kinetic data (joint moments, ground reaction forces (GRFs) and joint powers) for each single step of both legs. Unfiltered and filtered data are included. C3D files with raw marker and GRF data were recorded in Nexus (Vicon software, version 2.8.1) and are available upon request. Raw data were processed in Matlab (Mathworks 2016), including quality check, step determination and the exportation of data to .xls files. For each adult and for each walking speed, an .xls file was created, containing spatiotemporal parameters, medio-lateral (ML) and back-forward (BF) margins of stability (MoS), 3D joint angles, anterior-posterior (AP) and vertical GRFs, 3D joint moments and 3D joint power of each step of both legs. Overview files per walking speed condition are created in .xls, presenting the averaged gait parameters (calculated as average over all valid steps) of every subject. The processed data is also presented and visualized per gender for different age groups (18-29 years, 30-39 years, 40-49 years, 50-59 years, 60-69 years, ≥70 years). This can serve as normative data for treadmill based 3D gait analyses in adults, applicable for clinical and research purposes. Data is available at OSF.io (https://osf.io/t72cw/).

Influence of a passive exoskeleton on kinematics, joint moments, and self-reported ratings during a lifting task.

It was found that the Auxivo LiftSuit reduced the load on the back and hip muscles when lifting heavy loads, but its effect on lower body kinematics, joint moments, and self-reported ratings was unclear. The purpose of this study was to assess the effect of this passive lift-exoskeleton design, on lower body kinematics, joint moments, and self-reported ratings during lifting of heavy loads. Twenty healthy subjects performed lifting of heavy loads with and without the exoskeleton under surveillance of a motion capture system. Medium and maximum level adjustments of the exoskeleton, as well as no exoskeleton use were analyzed. Our results indicate significant reduction (p <.01) in pelvis segment tilt and hip flexion ROM with the exoskeleton at maximum level adjustment in males during lifting. Lumbosacral flexion moment ranges were significantly decreased (p <.013) with the exoskeleton at maximum and medium level adjustment in males during lifting. The general user impressions were mostly positive, with participants reporting that it was easier to perform the task with the exoskeleton than without it (p <.0.001), and preferring and recommending the exoskeleton for the task. Although our findings may suggest negative effects of the Auxivo LiftSuit in males and females due to a ROM restriction and loose fit, respectively, it does not mean that the Auxivo LiftSuit is not useful for lifting tasks. Further design improvements are required to allow full range of motion of hips and pelvis, as well to provide better adjustment and level of support in female users.

Elbow joint loads during simulated activities of daily living: implications for formulating recommendations after total elbow arthroplasty.

BACKGROUND: Overloading of the elbow joint prosthesis following total elbow arthroplasty can lead to implant failure. Joint moments during daily activities are not well contextualized for a prosthesis's failure limits, and the effect of the current postoperative instruction on elbow joint loading is unclear. This study investigates the difference in elbow joint moments between simulated daily tasks and between flexion-extension, pronation-supination, and varus-valgus movement directions. Additionally, the effect of the current postoperative instruction on elbow joint load is examined. METHODS: Nine healthy participants (age 45.8 ± 17 years, 3 males) performed 8 tasks; driving a car, opening a door, rising from a chair, lifting, sliding, combing hair, drinking, emptying cup, without and with the instruction "not lifting more than 1 kg." Upper limb kinematics and hand contact forces were measured. Elbow joint angles and net moments were analyzed using inverse dynamic analysis, where the net moments are estimated from movement data and external forces. RESULTS: Peak elbow joint moments differed significantly between tasks (P < .01) and movement directions (P < .01). The most and least demanding tasks were, rising from a chair (13.4 Nm extension, 5.0 Nm supination, and 15.2 Nm valgus) and sliding (4.3 Nm flexion, 1.7 Nm supination, and 2.6 Nm varus). Net moments were significantly reduced after instruction only in the chair task (P < .01). CONCLUSION: This study analyzed elbow joint moments in different directions during daily tasks. The outcomes question whether postoperative instruction can lead to decreasing elbow loads. Future research might focus on reducing elbow loads in the flexion-extension and varus-valgus directions.

Dual tasking increases kinematic and kinetic risk factors of ACL injury.

ACL injuries are common among athletes playing team sports. The impact of divided attention during team sports on landing mechanics is unclear. Twenty-one healthy females jumped at a 60° angle to their right and performed a second jump to their right or left at a 60° angle. The direction of the second jump was shown before movement (baseline) or mid-flight of the first jump (dual task). The signal for the dual-task conditions showed five arrows and the middle one indicated the jump direction (Flanker paradigm). The other arrows pointed in the same (congruent) or the opposite (incongruent) direction as the middle arrow. Results indicated larger initial and peak knee flexion angles, smaller peak knee valgus moments, and smaller vertical and posterior GRFs during baseline right jumps compared to other conditions. Peak posterior GRF was increased in the incongruent condition compared to the congruent condition during left jumps. Performance was decreased with longer stance times for the dual task compared to the baseline in both jump directions. Further, the incongruent condition had a longer stance time than the congruent condition during left jumps. More research focusing on decision-making with more challenging visual stimuli mimicking dynamic team sports is merited.

Prediction of Gait Kinematics and Kinetics: A Systematic Review of EMG and EEG Signal Use and Their Contribution to Prediction Accuracy.

Human-machine interfaces hold promise in enhancing rehabilitation by predicting and responding to subjects' movement intent. In gait rehabilitation, neural network architectures utilize lower-limb muscle and brain activity to predict continuous kinematics and kinetics during stepping and walking. This systematic review, spanning five databases, assessed 16 papers meeting inclusion criteria. Studies predicted lower-limb kinematics and kinetics using electroencephalograms (EEGs), electromyograms (EMGs), or a combination with kinematic data and anthropological parameters. Long short-term memory (LSTM) and convolutional neural network (CNN) tools demonstrated highest accuracies. EEG focused on joint angles, while EMG predicted moments and torque joints. Useful EEG electrode locations included C3, C4, Cz, P3, F4, and F8. Vastus Lateralis, Rectus Femoris, and Gastrocnemius were the most commonly accessed muscles for kinematic and kinetic prediction using EMGs. No studies combining EEGs and EMGs to predict lower-limb kinematics and kinetics during stepping or walking were found, suggesting a potential avenue for future development in this technology.

Lower Body Joint Moments during the Golf Swing in Older Adults: Comparison to Other Activities of Daily Living.

Golf participation has increased dramatically in the last several years. With this increase in participation, clinicians need better evidenced based strategies to advise those golfers with different pathologies when it is safe to return to the game. Golf teaching professionals also need to understand how to alter golf mechanics to protect injured and/or diseased joints in golfers to allow them to play pain free and avoid further injury. This study used a 3-dimensional link segment model to calculate the net joint moments on the large lower limb joints (knee and hip) during golf (lead and trail leg) and two commonly studied activities of daily living (gait and sit-to-stand) in 22 males, healthy, adult golfers. It also examined the correlations between these knee and hip joint loads and club head speed. The external valgus knee moment and the internal hip adduction moment were greater in the lead leg in golf than in the other activities and were also correlated with club head speed. This indicates a strategy of using the frontal plane GRF moment during the swing. The internal hip extension and knee flexion moment were also greater in the golf swing as compared with the other activities and the hip extension moment was also correlated with club head speed. This emphasizes the importance of hip extensor (i.e., gluteus maximus and hamstring) muscle function in golfers, especially in those emphasizing the use of anterior-posterior ground reaction forces (i.e., the pivoting moment). The golf swing places some loads on the knee and the hip that are much different than the loads during gait and sit-to-stand tasks. Knowledge of these golf swing loads can help both the clinician and golf professional provide better evidence-based advice to golfers in order to keep them healthy and avoid future pain/injury.

Performance of multiple neural networks in predicting lower limb joint moments using wearable sensors.

Joint moment measurements represent an objective biomechemical parameter in joint health assessment. Inverse dynamics based on 3D motion capture data is the current 'gold standard' to estimate joint moments. Recently, machine learning combined with data measured by wearable technologies such electromyography (EMG), inertial measurement units (IMU), and electrogoniometers (GON) has been used to enable fast, easy, and low-cost measurements of joint moments. This study investigates the ability of various deep neural networks to predict lower limb joint moments merely from IMU sensors. The performance of five different deep neural networks (InceptionTimePlus, eXplainable convolutional neural network (XCM), XCMplus, Recurrent neural network (RNNplus), and Time Series Transformer (TSTPlus)) were tested to predict hip, knee, ankle, and subtalar moments using acceleration and gyroscope measurements of four IMU sensors at the trunk, thigh, shank, and foot. Multiple locomotion modes were considered including level-ground walking, treadmill walking, stair ascent, stair descent, ramp ascent, and ramp descent. We show that XCM can accurately predict lower limb joint moments using data of only four IMUs with RMSE of 0.046 ± 0.013 Nm/kg compared to 0.064 ± 0.003 Nm/kg on average for the other architectures. We found that hip, knee, and ankle joint moments predictions had a comparable RMSE with an average of 0.069 Nm/kg, while subtalar joint moments had the lowest RMSE of 0.033 Nm/kg. The real-time feedback that can be derived from the proposed method can be highly valuable for sports scientists and physiotherapists to gain insights into biomechanics, technique, and form to develop personalized training and rehabilitation programs.

Machine learning-based prediction of joint moments based on kinematics in patients with cerebral palsy.

Joint moments during gait provide valuable information for clinical decision-making in patients with cerebral palsy (CP). Joint moments are calculated based on ground reaction forces (GRF) using inverse dynamics models. Obtaining GRF from patients with CP is challenging. Typically developed (TD) individuals' joint moments were predicted from joint angles using machine learning, but no such study has been conducted on patients with CP. Accordingly, we aimed to predict the dorsi-plantar flexion, knee flexion-extension, hip flexion-extension, and hip adduction-abduction moments based on the trunk, pelvis, hip, knee, and ankle kinematics during gait in patients with CP and TD individuals using one-dimensional convolutional neural networks (CNN). The anonymized retrospective gait data of 329 TD (26 years ± 14, mass: 70 kg ± 15, height: 167 cm ± 89) and 917 CP (17 years ± 9, mass:47 kg ± 19, height:153 cm ± 36) individuals were evaluated and after applying inclusion-exclusion criteria, 132 TD and 622 CP patients with spastic diplegia were selected. We trained specific CNN models and evaluated their performance using isolated test subject groups based on normalized root mean square error (nRMSE) and Pearson correlation coefficient (PCC). Joint moments were predicted with nRMSE between 18.02% and 13.58% for the CP and between 12.55% and 8.58% for the TD groups, whereas with PCC between 0.85 and 0.93 for the CP and between 0.94 and 0.98 for the TD groups. Machine learning-based joint moment prediction from kinematics could replace conventional moment calculation in CP patients in the future, but the current level of prediction errors restricts its use for clinical decision-making today.

Enforcing walking speed and step-length affects joint kinematics and kinetics in male and female healthy adults.

BACKGROUND: Individuals increase walking speed by increasing their step-length, increasing their step-frequency, or both. During basic training military recruits are introduced to marching "in-step", and thus the requirement to walk at fixed speeds and step-lengths. The extent to which individuals are required to under- or over-stride will vary depending on their stature, and the stature of others in their section. The incidence of stress fractures in female recruits undergoing basic training is higher than that for their male counterparts. RESEARCH QUESTION: Therefore, the purpose of this study was to determine how joint kinematics and kinetics are affected by walking speed, step-length, and sex. METHODS: Thirty-seven (19 female) aerobically active non-injured individuals volunteered for this study. Synchronised three-dimensional kinematic and kinetic data were collected while participants walked overground at prescribed speeds. Audio and visual cues were used to control step-lengths. Linear mixed models were run to analyse the effects of speed, step-length condition, and sex on peak joint moments. RESULTS AND SIGNIFICANCE: The findings of this study showed that, in general, walking faster and over-striding predominantly increased peak joint moments, suggesting that over-striding is more likely to negatively affect injury risk than under-striding. This is especially important for individuals unaccustomed to over-striding as the cumulative effect of increased joint moments may affect a muscles capability to withstand the increased external forces associated with walking faster and with longer step-lengths, which could then lead to an increased risk of developing an injury.

Peak Tibiofemoral Contact Forces Estimated Using IMU-Based Approaches Are Not Significantly Different from Motion Capture-Based Estimations in Patients with Knee Osteoarthritis.

Altered tibiofemoral contact forces represent a risk factor for osteoarthritis onset and progression, making optimization of the knee force distribution a target of treatment strategies. Musculoskeletal model-based simulations are a state-of-the-art method to estimate joint contact forces, but they typically require laboratory-based input and skilled operators. To overcome these limitations, ambulatory methods, relying on inertial measurement units, have been proposed to estimated ground reaction forces and, consequently, knee contact forces out-of-the-lab. This study proposes the use of a full inertial-capture-based musculoskeletal modelling workflow with an underlying probabilistic principal component analysis model trained on 1787 gait cycles in patients with knee osteoarthritis. As validation, five patients with knee osteoarthritis were instrumented with 17 inertial measurement units and 76 opto-reflective markers. Participants performed multiple overground walking trials while motion and inertial capture methods were synchronously recorded. Moderate to strong correlations were found for the inertial capture-based knee contact forces compared to motion capture with root mean square error between 0.15 and 0.40 of body weight. The results show that our workflow can inform and potentially assist clinical practitioners to monitor knee joint loading in physical therapy sessions and eventually assess long-term therapeutic effects in a clinical context.

Allometric Scaling Hip Joint Moments Optimally Reduces Anthropometric Differences in Males and Females.

Running biomechanics are scaled to reduce the effects of anthropometric differences between participants. Ratio scaling has limitations, and allometric scaling has not been applied to hip joint moments. The aim was to compare raw, ratio and allometrically scaled hip joint moments. Sagittal and frontal plane moments of 84 males and 47 females were calculated while running at 4.0 m/s. Raw data were ratio scaled by body mass (BM), height (HT), leg length (LL) and BM multiplied by HT (BM*HT) and LL (BM*LL). Log-linear (for BM, HT and LL individually) or log-multilinear regression (BM*HT and BM*LL) exponents were calculated. Correlations and r2 values assessed the effectiveness of each scaling method. Eighty-five per cent of raw moments were positively correlated to the anthropometrics with r2 values of 10-19%. In ratio scaling, 26-43% were significantly correlated to the moments and a majority were negative, indicating overcorrections. The most effective scaling procedure was the allometric BM*HT, as the mean shared variance between the hip moment and anthropometrics was 0.1-0.2% across all sexes and moments and none had significant correlations. Allometric scaling of hip joint moments during running are advised if the goal is to remove the underlying effects of anthropometrics across male and female participants.

Dataset of 3D gait analysis in typically developing children walking at three different speeds on an instrumented treadmill in virtual reality.

In this article, gait data of typically developing (TD) children (24 boys/31 girls, mean (95% confidence interval) age 9.38 (8.51 - 10.25) years, body mass 35.67 (31.40 - 39.94) kg, leg length 0.73 (0.70 - 0.76) m, and height 1.41 (1.35 - 1.46) m) walking at different walking speeds is shared publicly. Raw and processed data is presented for each child separately and includes data of each single step of both legs. Beside, the subject demographics and the results from the physical examination are presented allowing to select TD children from the database to create a matched group, based on specific parameters (e.g. sex and body weight). For clinical application, gait data is also presented per age group, which provides quick insight into the normal gait pattern of TD children of varying age. Gait analysis was performed during treadmill walking in a virtual environment using the Computer Assisted Rehabilitation Environment (CAREN). The human body lower limb model with trunk markers (HBM2) was used as biomechanical model. Children walked at comfortable walking speed, 30% slower and 30% faster (random sequence) while wearing gymnastic shoes and a safety harness to prevent falling. For each speed condition, 250 steps were recorded. Data quality check, step detection and the calculation of gait parameters was done by custom made Matlab algorithms. Raw data files are provided per walking speed, for each child separately. The raw data is exported from the CAREN software (D-flow) and is provided in .mox and .txt files. It includes the output from the models such as subject data, marker and force data, kinematic data (joint angles), kinetic data (joint moments, GRFs, joint powers), as well as CoM data and EMG data (the last two are not described in this manuscript), for each speed condition and each child. Unfiltered and filtered data are included. C3D files with raw marker and GRF data were recorded in Nexus (Vicon software) and are available upon request. After analyzing the raw data into Matlab (R2016a, Mathworks) using custom made Matlab algorithms, processed data is obtained. The processed data is provided in .xls files and is also presented for each child separately. It contains spatiotemporal parameters, 3D joint angles, anterior-posterior and vertical ground reaction forces (GRF), 3D joint moments and sagittal joint power of each step of the left and right leg. In addition to each individual's data, overview files (.xls) are created per walking speed condition. These overviews present the averaged gait parameter (e.g. joint angle), calculated over all valid steps, of each child.

The effect of trunk and shank position on the hip-to-knee moment ratio in a bilateral squat.

OBJECTIVES: The effect of knee position on joint moments during squats has been studied; however, the effect of trunk angle has been less well investigated. This study evaluated the effect of both trunk and knee sagittal plane position on the distribution of moments between the hip and knee extensors during the bilateral squat. DESIGN: Observational study. SETTING: Biomechanics laboratory. PARTICIPANTS: One hundred individuals performed bilateral squats. MAIN OUTCOME MEASURES: Motion and force data were collected using motion capture and force plates. Trunk and shank angles and hip and knee moments were calculated. A linear regression was used to associate the ratio between the hip and knee moments (hip-to-knee moment ratio) with the sagittal plane trunk and shank angles, while accounting for six squat depths (between 60° and 110° of knee flexion) and side. RESULTS: Trunk angle and shank angle each contributed to the hip-to-knee moment ratio (P < .001) with trunk accounting for a higher proportion of variance than the other variables. The hip-to-knee moment ratio increased with increasing trunk angle and with decreasing shank angle. CONCLUSIONS: This large cohort study supports the use of trunk position to instruct squat technique with the goal of modifying hip and knee moments.

Transversal Malalignment and Proximal Involvement Play a Relevant Role in Unilateral Cerebral Palsy Regardless the Subtype.

Classification of gait disorders in cerebral palsy (CP) remains challenging. The Winters, Gage, and Hicks (WGH) is a commonly used classification system for unilateral CP regarding the gait patterns (lower limb kinematics) solely in the sagittal plane. Due to the high number of unclassified patients, this classification system might fail to depict all gait disorders accurately. As the information on trunk/pelvic movements, frontal and transverse planes, and kinetics are disregarded in WGH, 3D instrumented gait analysis (IGA) for further characterization is necessary. The objective of this study was a detailed analysis of patients with unilateral CP using IGA taking all planes/degrees of freedom into account including pelvic and trunk movements. A total of 89 individuals with unilateral CP matched the inclusion criteria and were classified by WGH. Subtype-specific differences were analyzed. The most remarkable findings, in addition to the established WGH subtype-specific deviations, were pelvic obliquity and pelvic retraction in all WGH types. Furthermore, the unclassified individuals showed altered hip rotation moments and pelvic retraction almost throughout the whole gait cycle. Transversal malalignment and proximal involvement are relevant in all individuals with unilateral CP. Further studies should focus on WGH type-specific rotational malalignment assessment (static vs. dynamic, femoral vs. tibial) including therapeutic effects and potential subtype-specific compensation mechanisms and/or tertiary deviations of the sound limb.

Quantifying the hip-ankle synergy in short-term maximal cycling.

Simulation studies have demonstrated that the hip and ankle joints form a task-specific synergy during the downstroke in maximal cycling to enable the power produced by the hip extensor muscles to be transferred to the crank. The existence of the hip-ankle synergy has not been investigated experimentally. Therefore, we sought to apply a modified vector coding technique to quantify the strength of the hip-ankle moment synergy in the downstroke during short-term maximal cycling at a pedalling rate of 135 rpm. Twelve track sprint cyclists performed 3 × 4 s seated sprints at 135 rpm, interspersed with 2 × 4 s seated sprints at 60 rpm on an isokinetic ergometer. Data from the 60 rpm sprints were not analysed in this study. Joint moments were calculated via inverse dynamics, using pedal forces and limb kinematics. The hip-ankle moment synergy was quantified using a modified vector coding method. Results showed, for 28.8% of the downstroke the hip and ankle moments were in-phase, demonstrating the hip and ankle joints tend to work in synergy in the downstroke, providing some support findings from simulation studies of cycling. At a pedalling rate of 135 rpm the hip-phase was most frequent (42.5%) significantly differing from the in- (P = 0.044), anti- (P < 0.001), and ankle-phases (P = 0.004), demonstrating hip-dominant action. We believe this method shows promise to answer research questions on the relative strength of the hip-ankle synergy between different cycling conditions (e.g., power output and pedalling rates).

Estimation of Joint Moments During Turning Maneuvers in Alpine Skiing Using a Three Dimensional Musculoskeletal Skier Model and a Forward Dynamics Optimization Framework.

In alpine skiing, estimation of the joint moments acting onto the skier is essential to quantify the loading of the skier during turning maneuvers. In the present study, a novel forward dynamics optimization framework is presented to estimate the joint moments acting onto the skier incorporating a three dimensional musculoskeletal model (53 kinematic degrees of freedom, 94 muscles). Kinematic data of a professional skier performing a turning maneuver were captured and used as input data to the optimization framework. In the optimization framework, the musculoskeletal model of the skier was applied to track the experimental data of a skier and to estimate the underlying joint moments of the skier at the hip, knee and ankle joints of the outside and inside leg as well as the lumbar joint. During the turning maneuver the speed of the skier was about 14 m/s with a minimum turn radius of about 16 m. The highest joint moments were observed at the lumbar joint with a maximum of 1.88 Nm/kg for lumbar extension. At the outside leg, the highest joint moments corresponded to the hip extension moment with 1.27 Nm/kg, the knee extension moment with 1.02 Nm/kg and the ankle plantarflexion moment with 0.85 Nm/kg. Compared to the classical inverse dynamics analysis, the present framework has four major advantages. First, using a forward dynamic optimization framework the underlying kinematics of the skier as well as the corresponding ground reaction forces are dynamically consistent. Second, the present framework can cope with incomplete data (i.e., without ground reaction force data). Third, the computation of the joint moments is less sensitive to errors in the measurement data. Fourth, the computed joint moments are constrained to stay within the physiological limits defined by the musculoskeletal model.

Comparison of sagittal plane gait characteristics between the overground and treadmill approach for gait analysis in typically developing children.

Background: Instrumented treadmills have become more mainstream in clinical assessment of gait disorders in children, and are increasingly being applied as an alternative to overground gait analysis. Both approaches differ in multiple elements of set-up (e.g., overground versus treadmill, Pug-in Gait versus Human Body Model-II), workflow (e.g., limited amount of steps versus many successive steps) and post-processing of data (e.g., different filter techniques). These individual elements have shown to affect gait. Since the approaches are used in parallel in clinical practice, insight into the compound effect of the multiple different elements on gait is essential. This study investigates whether the outcomes of two approaches for 3D gait analysis are interchangeable in typically developing children. Methods: Spatiotemporal parameters, sagittal joint angles and moments, and ground reaction forces were measured in typically developing children aged 3-17 years using the overground (overground walking, conventional lab environment, Plug-In Gait) and treadmill (treadmill walking in virtual environment, Human Body Model-II) approach. Spatiotemporal and coefficient of variation parameters, and peak values in kinematics and kinetics of both approaches were compared using repeated measures tests. Kinematic and kinetic waveforms from both approaches were compared using statistical parametric mapping (SPM). Differences were quantified by mean differences and root mean square differences. Results: Children walked slower, with lower stride and stance time and shorter and wider steps with the treadmill approach than with the overground approach. Mean differences ranged from 0.02 s for stride time to 3.3 cm for step width. The patterns of sagittal kinematic and kinetic waveforms were equivalent for both approaches, but significant differences were found in amplitude. Overall, the peak joint angles were larger during the treadmill approach, showing mean differences ranging from 0.84° (pelvic tilt) to 6.42° (peak knee flexion during swing). Mean difference in peak moments ranged from 0.02 Nm/kg (peak knee extension moment) to 0.32 Nm/kg (peak hip extension moment), showing overall decreased joint moments with the treadmill approach. Normalised ground reaction forces showed mean differences ranging from 0.001 to 0.024. Conclusion: The overground and treadmill approach to 3D gait analysis yield different sagittal gait characteristics. The systematic differences can be due to important changes in the neuromechanics of gait and to methodological choices used in both approaches, such as the biomechanical model or the walkway versus treadmill. The overview of small differences presented in this study is essential to correctly interpret the results and needs to be taken into account when data is interchanged between approaches. Together with the research/clinical question and the context of the child, the insight gained can be used to determine the best approach.

Predicting the Internal Knee Abduction Impulse During Walking Using Deep Learning.

Knee joint moments are commonly calculated to provide an indirect measure of knee joint loads. A shortcoming of inverse dynamics approaches is that the process of collecting and processing human motion data can be time-consuming. This study aimed to benchmark five different deep learning methods in using walking segment kinematics for predicting internal knee abduction impulse during walking. Three-dimensional kinematic and kinetic data used for the present analyses came from a publicly available dataset on walking (participants n = 33). The outcome for prediction was the internal knee abduction impulse over the stance phase. Three-dimensional (3D) angular and linear displacement, velocity, and acceleration of the seven lower body segment's center of mass (COM), relative to a fixed global coordinate system were derived and formed the predictor space (126 time-series predictors). The total number of observations in the dataset was 6,737. The datasets were split into training (75%, n = 5,052) and testing (25%, n = 1685) datasets. Five deep learning models were benchmarked against inverse dynamics in quantifying knee abduction impulse. A baseline 2D convolutional network model achieved a mean absolute percentage error (MAPE) of 10.80%. Transfer learning with InceptionTime was the best performing model, achieving the best MAPE of 8.28%. Encoding the time-series as images then using a 2D convolutional model performed worse than the baseline model with a MAPE of 16.17%. Time-series based deep learning models were superior to an image-based method when predicting knee abduction moment impulse during walking. Future studies looking to develop wearable technologies will benefit from knowing the optimal network architecture, and the benefit of transfer learning for predicting joint moments.

Knee Arthrodesis Affects Gait Kinematics More in the Ankle Than in the Hip Joint.

Background and Objectives: No gold standard exists for treating persistent periprosthetic knee infections. Knee arthrodesis represents one treatment concept for extensive bone defects and extensor system insufficiencies. It has already been shown that knee arthrodesis leads to a significant reduction in one's quality of life. The aim of this survey was to assess the influence of knee arthrodesis on the neighboring joints on the basis of gait analysis data. Our hypothesis is that the hip and ankle joints are negatively influenced by knee arthrodesis in the process of walking. Materials and methods: We performed six pedobarographic and four gait analytical measurements in six patients 2.4 ± 1.6 years after receiving knee arthrodesis at the operating ages of 69.1 ± 9.2 years. Gait analysis consisted of time-distance parameters/minute (number of steps, double support, cycle time, standing phase, step length, gait speed). A healthy group of test subjects (n = 52) was included as the control cohort. Gait analysis was conducted using a three-dimensional movement system and three force-measuring platforms to determine the ground reaction force. Foot pressure was measured using a pedography platform. Results: Five of six patients presented an incomplete rolling movement over the toes on the side that was operated on, presenting with a gait line ending in the forefoot area. All of the patients bore less weight on the side that was operated on. Three of six patients demonstrated a pathological gait line with a healthy opposite side ending in the forefoot area. All of the patients exhibited a reduction in gait speed and step length and a lower number of steps. All of the patients had a prolonged double support/cycle time. Conclusions: Isolated knee arthrodesis is associated with reduced forefoot repulsion, restricted movement on the side receiving the operation, and reduced movement in the ankle/knee joint. The hip showed norm deviations in the hip moment/angle. Knee arthrodesis causes reduced gait kinetics/kinematics. Our survey shows that the relative joint moments of the ankle joint and hip are often reduced. The ankle joint is more affected compared to the hip.