Three-dimensional Ground Reaction Forces Research Articles

Impact of Leg Strength on the Force Produced During a Cross-Punch in Highly Trained Amateur Boxers.

Satkunskiene, D, Bruzas, V, Mantas, M, Snieckus, A, and Kamandulis, S. Impact of leg strength on the force produced during a cross-punch in highly trained amateur boxers leg strength impact on cross-punch force. J Strength Cond Res XX(X): 000-000, 2024-The purpose of this study was to determine the impact of leg strength on force generation during the execution of a cross-punch by highly trained amateur boxers. This study involved 12 male, amateur boxers with a body mass range of 64.5-112.2 kg and a mean of 89.2 ± 17.0 kg. This study used an isokinetic dynamometer to measure the peak torque of the knee extensor and flexor muscles at 60°/s. The force plate was used to collect three-dimensional ground reaction forces (GRF), moments (GRM), and the center of pressure (COP) during a maximal cross-punch. The Kiktest-100 system was used to determine the force of the punch, while the Hykso punch tracker accelerometer measured the maximum fist velocity. This study revealed a positive correlation between knee extensor strength and GRFz, GRFy, and COPy during punch performance. In addition, the GRMz gradient and GRMz impulse during the execution phase have a significant positive effect on cross-punch force. In conclusion, the force of a straight punch in highly trained, amateur boxers is related to a greater angular velocity about the vertical axis during body rotation.

Markerless motion capture provides accurate predictions of ground reaction forces across a range of movement tasks

Measuring or estimating the forces acting on the human body during movement is critical for determining the biomechanical aspects relating to injury, disease and healthy ageing. In this study we examined whether quantifying whole-body motion (segmental accelerations) using a commercial markerless motion capture system could accurately predict three-dimensional ground reaction force during a diverse range of human movements: walking, running, jumping and cutting. We synchronously recorded 3D ground reaction forces (force instrumented treadmill or in-ground plates) with high-resolution video from eight cameras that were spatially calibrated relative to a common coordinate system. We used a commercially available software to reconstruct whole body motion, along with a geometric skeletal model to calculate the acceleration of each segment and hence the whole-body centre of mass and ground reaction force across each movement task. The average root mean square difference (RMSD) across all three dimensions and all tasks was 0.75 N/kg, with the maximum average RMSD being 1.85 N/kg for running vertical force (7.89 % of maximum). There was very strong agreement between peak forces across tasks, with R2 values indicating that the markerless prediction algorithm was able to predict approximately 95–99 % of the variance in peak force across all axes and movements. The results were comparable to previous reports using whole-body marker-based approaches and hence this provides strong proof-of-principle evidence that markerless motion capture can be used to predict ground reaction forces and therefore potentially assess movement kinetics with limited requirements for participant preparation.

Evaluation of Resultant Ground Reaction Force and Its Velocity Using the Elapsed Time and Peak Reaction Force at 3-Dimensional Direction During Landing Task

The purpose of this study was to analyze the effects of bilateral landing leg and sex on resultant ground reaction force (GRF) and its velocity, using peak vertical GRF and elapsed time during drop landing tasks. A repeated-measures two-way analysis of variance explored the impact of landing legs and sex on the resultant vector, three-dimensional GRF, elapsed time, and velocity in 40 participants (20 males and 20 females). Participants performed drop landings from a 35-cm box. Effects of sex and landing leg were analyzed using a repeated-measures model (two sexes × two legs) on GRF magnitude and velocity. Females displayed shorter elapsed times to peak GRF compared to males in the anterior-posterior and vertical directions. Significant differences emerged between sexes in both magnitude and velocity of resultant and peak vertical GRF, with females exhibiting higher values. This suggests the adoption of distinct landing strategies between sexes. Notably, no significant differences were found in GRF magnitude or velocity between bilateral leg landings. These results indicate that healthy individuals of both sexes utilize different landing strategies during drop landings. This knowledge has potential applications in clinical settings for evaluating impulse force and stress transfer to the musculoskeletal system during landing tasks.

Estimation of Three-Dimensional Ground Reaction Forces During Walking and Turning Using Insole Pressure Sensors Based on Gait Pattern Recognition

Estimation of Three-Dimensional Ground Reaction Forces During Walking and Turning Using Insole Pressure Sensors Based on Gait Pattern Recognition

The Impact of Excessive Body Weight and Foot Pronation on Running Kinetics: A Cross-Sectional Study

BackgroundRunning exercise is an effective means to enhance cardiorespiratory fitness and body composition. Besides these health benefits, running is also associated with musculoskeletal injuries that can be more prevalent in individuals with excessive body weight. Little is known regarding the specific effects of overweight and foot pronation on ground reaction force distribution during running. Therefore, this study aimed to investigate the effects of overweight/obesity and foot pronation on running kinetics.MethodsEighty-four young adults were allocated to four experimental groups: non-excessive body weight/non-pronated feet; non-excessive body weight/pronated feet; overweight or obesity/ non-pronated feet and overweight or obesity/pronated feet. Biomechanical testing included participants to run at ~ 3.2 m/s over an 18-m walkway with an embedded force plate at its midpoint. Three-dimensional ground reaction forces were recorded and normalized to body mass to evaluate running kinetics from 20 running trials. Test–re-test reliability for running speed data demonstrated ICC > 0.94 for each group and in total.ResultsThe results indicated significantly lower vertical impact peak forces (p = 0.001, effect size = 0.12), shorter time to reach the vertical impact peak (p = 0.006, effect size = 0.08) and reduced vertical loading rate (p = 0.0007, effect size = 0.13) in individuals with excessive body weight (overweight or obesity/non-pronated feet group and overweight or obesity/pronated feet) compared with individuals non-excessive body weight (non-excessive body weight/non-pronated feet and non-excessive body weight/pronated feet). Moreover, the excessive body weight groups presented lower peak braking (p = 0.01, effect size = 0.06) and propulsion forces (p = 0.003, effect size = 0.09), lower medio-lateral loading rate (p = 0.0009, effect size = 0.12), and greater free moments (p = 0.01, effect size = 0.07) when compared to the non-overweight groups. Moreover, a significant body mass by foot pronation interaction was found for peak medio-lateral loading rate. Non-excessive body weight/pronated feet, excessive body weight/non-pronated feet and excessive body weight/pronation groups presented lower medio-lateral loading rates compared to non-excessive body weight/non-pronated feet (p = 0.0001, effect size = 0.13).ConclusionsOur results suggest that excessive body weight has an impact on ground reaction forces during running. We particularly noted an increase in medio-lateral and torsional forces during the stance phase. Individuals with excessive body weight appear to adapt their running patterns in an effort to attenuate early vertical impact loading.

Identification of subject-specific responses to footwear during running

Placing a stronger focus on subject-specific responses to footwear may lead to a better functional understanding of footwear’s effect on running and its influence on comfort perception, performance, and pathogenesis of injuries. We investigated subject-specific responses to different footwear conditions within ground reaction force (GRF) data during running using a machine learning-based approach. We conducted our investigation in three steps, guided by the following hypotheses: (I) For each subject x footwear combination, unique GRF patterns can be identified. (II) For each subject, unique GRF characteristics can be identified across footwear conditions. (III) For each footwear condition, unique GRF characteristics can be identified across subjects. Thirty male subjects ran ten times at their preferred (self-selected) speed on a level and approximately 15 m long runway in four footwear conditions (barefoot and three standardised running shoes). We recorded three-dimensional GRFs for one right-foot stance phase per running trial and classified the GRFs using support vector machines. The highest median prediction accuracy of 96.2% was found for the subject x footwear classification (hypothesis I). Across footwear conditions, subjects could be discriminated with a median prediction accuracy of 80.0%. Across subjects, footwear conditions could be discriminated with a median prediction accuracy of 87.8%. Our results suggest that, during running, responses to footwear are unique to each subject and footwear design. As a result, considering subject-specific responses can contribute to a more differentiated functional understanding of footwear effects. Incorporating holistic analyses of biomechanical data is auspicious for the evaluation of (subject-specific) footwear effects, as unique interactions between subjects and footwear manifest in versatile ways. The applied machine learning methods have demonstrated their great potential to fathom subject-specific responses when evaluating and recommending footwear.

The effect of severe obesity on three-dimensional ground reaction force signals during walking

BackgroundThe gait pattern of adults with class I obesity [30 ≤ body mass index < 35kg/m2] was characterized by altered three-dimensional ground reaction force signals compared to lean adults (18.5 ≤ body mass index < 25 kg/m2). However, results might not be generalizable to adults with severe obesity (class II and III; body mass index ≥ 35 kg/m2). Hence, the purpose of the present study was to investigate the differences in relative ground reaction force signals, i.e., normalized by body weight, between adults with severe obesity and lean adults using functional principal component analysis. MethodsThirteen lean and eighteen sedentary adults with severe obesity performed a 5-min walking trial (1.11 m/s) on an instrumented treadmill. The first five functional principal components of the relative force signals (mediolateral, anterior-posterior, and vertical directions) were obtained using functional principal component analysis. Functional principal component scores were compared between groups using an analysis of covariance with age as covariable. FindingsFunctional principal component analysis reported a statistically significant group effect for first functional principal component score for mediolateral (P = 0.004), and second and fifth functional principal component scores for anterior-posterior (P ≤ 0.02) force signals. Adults with severe obesity displayed a greater mediolateral force during most of the stance but similar magnitudes of the anterior-posterior and vertical forces compared to lean adults. InterpretationTherefore, increasing the obesity level accentuates differences in mediolateral force but promotes no specific changes in anterior-posterior force likely due to chronic loading adaptation.

The Relationship between Ground Reaction Forces, Foot Positions and Type of Clubs Used in Golf: A Systematic Review and Meta-Analysis

Background: Despite the growing interest in golf, the impact of motion control strategy on golf performance may be uncertain. A network meta-analysis was conducted to investigate the relationships between ground reaction force (GRF), foot positions, and club types in golf and determine whether different clubs have a different impact when swinging. Methods: Three databases were searched from the time of inception to March 2023. Eligibility criteria included studies that provided data using a driver to swing and compared outcomes to control data. Results: Searches yielded 6527 studies. Seven studies met the selection criteria (n = 422 participants). The results showed that an iron 6 is the best when considering the trail foot vertical ground reaction force (TVGRF), trail foot anteroposterior ground reaction force (TAGRF), and lead foot vertical ground reaction force (LVGRF). The pitching wedge was the best in the lead foot mediolateral ground reaction force (LMGRF) and lead foot anteroposterior ground reaction force (LAGRF). Iron 7 was the best in the trail foot mediolateral ground reaction force (TMGRF), and the lead foot was larger than the trail foot to the vertical GRF. Discussion: The study found that clubs may influence a player’s posture and swing power because golf clubs are available in various lengths and shapes. The lead foot generates a larger GRF than the trail foot; three-dimensional GRFs differ among golf clubs. When a golfer aims to maximize the distance of their drives, they must generate relatively more resultant horizontal reaction force (RFH). Golfers often use different clubs to achieve optimal performance on the course by controlling their motion. However, there needs to be a focus on the quality of the included studies because the sample size was too small, increasing the risk of bias associated with the results.

Force-plate derived predictors of lateral jump performance in NCAA Division-I men's basketball players.

A lateral jump assessment may provide unique benefits in sports such as basketball that require multidirectional performance optimization. This study aimed to examine selected force-plate derived metrics as predictors of lateral jump task distance in men's basketball players. Twenty-two NCAA Division-I men's basketball players (19.4 ± 1.3 years, 95.0 ± 12.5 kg, 196.5 ± 8.1 cm) each performed six single leg lateral jumps while standing on a force plate (1200 Hz, Kistler Instrument Corp). The lateral jump task involved the subject beginning by standing on the force plate and jumping sideways off one foot and then landing on the floor with the opposite foot. Three-dimensional ground reaction force curves were used to identify the eccentric and concentric phases of the jump and variables were computed each from the lateral (y), vertical (z), and resultant (r) force traces. Peak ground reaction force (pGRF), ground reaction force angle (θr), eccentric braking rate of force development (ECC-RFD), average concentric force (CON-AVG), total jump duration, eccentric phase duration, and eccentric to total time ratio were evaluated for predictive ability. Three regression models were able to significantly (p<0.05) predict jump distance: (1) pGRFy, pGRFz, and θr (p<0.001, R2 = 0.273), (2) Relative pGRFy, Relative pGRFz, and θr ((p<0.001, R2 = 0.214), and (3) Relative CON-AVGy and Relative pGRFr (p<0.001, R2 = 0.552). While several force plate-derived metrics were identified as significant predictors, a model with Relative CON-AVGy and Relative pGRFr explained a greater variability in performance (R2 = 0.55) compared to the other variables which were low, yet also significant. These results suggest that lateral ground reaction forces can be used to evaluate lateral jump performance with the use of three-dimensional force plates. The identified predictors can be used as a starting point for performance monitoring, as basketball training interventions can be directed at specific improvements in the identified metrics.

Automated recognition of asymmetric gait and fatigue gait using ground reaction force data.

Introduction: The purpose of this study was to evaluate the effect of running-induced fatigue on the characteristic asymmetry of running gait and to identify non-linear differences in bilateral lower limbs and fatigued gait by building a machine learning model. Methods: Data on bilateral lower limb three-dimensional ground reaction forces were collected from 14 male amateur runners before and after a running-induced fatigue experiment. The symmetry function (SF) was used to assess the degree of symmetry of running gait. Statistical parameter mapping (Paired sample T-test) algorithm was used to examine bilateral lower limb differences and asymmetry changes pre- and post-fatigue of time series data. The support vector ma-chine (SVM) algorithm was used to recognize the gait characteristics of both lower limbs before and after fatigue and to build the optimal algorithm model by setting different kernel functions. Results: The results showed that the ground reaction forces were asymmetrical (SF > 0.5) both pre-and post-fatigue and mainly concentrated in the medial-lateral direction. The asymmetry of the medial-lateral direction increased significantly after fatigue (p < 0.05). In addition, we concluded that the polynomial kernel function could make the SVM model the most accurate in classifying left and right gait features (accuracy of 85.3%, 82.4%, and 82.4% in medial-lateral, anterior-posterior and vertical directions, respectively). Gaussian radial basis kernel function was the optimal kernel function of the SVM algorithm model for fatigue gait recognition in the medial-lateral and vertical directions (accuracy of 54.2% and 62.5%, respectively). Moreover, polynomial was the optimal kernel function of the anterior-posterior di-rection (accuracy = 54.2%). Discussion: We proved in this study that the SVM algorithm model depicted good performance in identifying asymmetric and fatigue gaits. These findings can provide implications for running injury prevention, movement monitoring, and gait assessment.

Regression transfer learning for the prediction of three-dimensional ground reaction forces and joint moments during gait

Regression transfer learning for the prediction of three-dimensional ground reaction forces and joint moments during gait

Regression transfer learning for the prediction of three-dimensional ground reaction forces and joint moments during gait

Regression transfer learning for the prediction of three-dimensional ground reaction forces and joint moments during gait

Estimating Ground Reaction Forces from Two-Dimensional Pose Data: A Biomechanics-Based Comparison of AlphaPose, BlazePose, and OpenPose.

The adoption of computer vision pose estimation approaches, used to identify keypoint locations which are intended to reflect the necessary anatomical landmarks relied upon by biomechanists for musculoskeletal modelling, has gained increasing traction in recent years. This uptake has been further accelerated by keypoint use as inputs into machine learning models used to estimate biomechanical parameters such as ground reaction forces (GRFs) in the absence of instrumentation required for direct measurement. This study first aimed to investigate the keypoint detection rate of three open-source pose estimation models (AlphaPose, BlazePose, and OpenPose) across varying movements, camera views, and trial lengths. Second, this study aimed to assess the suitability and interchangeability of keypoints detected by each pose estimation model when used as inputs into machine learning models for the estimation of GRFs. The keypoint detection rate of BlazePose was distinctly lower than that of AlphaPose and OpenPose. All pose estimation models achieved a high keypoint detection rate at the centre of an image frame and a lower detection rate in the true sagittal plane camera field of view, compared with slightly anteriorly or posteriorly located quasi-sagittal plane camera views. The three-dimensional ground reaction force, instantaneous loading rate, and peak force for running could be estimated using the keypoints of all three pose estimation models. However, only AlphaPose and OpenPose keypoints could be used interchangeably with a machine learning model trained to estimate GRFs based on AlphaPose keypoints resulting in a high estimation accuracy when OpenPose keypoints were used as inputs and vice versa. The findings of this study highlight the need for further evaluation of computer vision-based pose estimation models for application in biomechanical human modelling, and the limitations of machine learning-based GRF estimation models that rely on 2D keypoints. This is of particular relevance given that machine learning models informing athlete monitoring guidelines are being developed for application related to athlete well-being.

Machine learning-based classification of healthy and impaired gaits using 3D-GRF signals

Gait analysis is helpful for rehabilitation, clinical diagnoses, and sporting activities. Among the gathered signals, ground reaction forces (GRF) may be used for assisting doctors in recognizing and categorizing gait patterns using Machine-Learning methods. In this study, GaitRec and Gutenberg databases were used, where GaitRec contains 2645 gait disorder (GD) patients and 211 Healthy Controls (HCs), and the Gutenberg database has 350 HCs. The combined database has HCs and four GD classes: hip, knee, ankle, and calcaneus. GD is an abnormality in the hip, knee, or ankle joints, whereas Calcaneus gait is calcaneus fractures or ankle fusions. We pre-processed the GRF signals, applied different feature extraction techniques, removed the highly correlated features, and ranked the features using three feature selection algorithms. K-nearest neighbour model (KNN) showed the top performance in terms of accuracy in all experiments. Four different experimental schemes were pursued: (i) 6 binary classifications; (ii) 1 three-class classification; (iii) 2 four-class classifications; (iv) one five-class classification. We also compared the performance of vertical GRF with three-dimensional GRF. We found that using three-dimensional GRF increased the overall performance. Furthermore, it is found that time-domain and Wavelet features are among the most useful in identifying gait patterns. The findings show promising performance in automated gait disorder classification.

Analysis of Running Gait in Children with Cerebral Palsy: Barefoot vs. a New Ankle Foot Orthosis.

Running is an essential activity for children with cerebral palsy (CP). This study aims to characterize the locomotor pattern of running in hemiplegic children with new generation ankle foot orthosis (AFOs) conceived to foster intense motor activities such as running. A group of 18 children with spastic hemiplegia was recruited. A biomechanical multivariable comparison was made between barefoot and with AFO running trials. The focus was devoted to bilateral sagittal plane hip, knee, ankle kinematics and kinetics, and three-dimensional ground reaction forces. Wearing the orthoses, the children were found to reduce cadence and the duration of the stance phase as well as increase the step and stride length. The new AFO resulted in significant changes in kinematics of affected ankle both at initial contact 0-3% GC (p < 0.017) and during the entire swing phase 31-100%GC (p < 0.001) being the ankle more dorsiflexed with AFO compared to barefoot condition. Ankle power was found to differ significantly both in absorption and generation 5-10%GC (p < 0.001); 21-27%GC (p < 0.001) with a reduction in both cases when the AFO was worn. No statistical differences were recorded in the GRF components, in the affected ankle torque and hip and knee kinematics and kinetics.

Treadmill vs. overground trotting - a comparison of two kinetic measurement systems

Force plates are considered the gold standard for kinetic gait analysis, benefiting from measuring three-dimensional ground reaction forces. Nevertheless, the major disadvantage is that many trials are required during overground locomotion to capture adequate single-limb contacts. Additionally, the dogs slightly change velocities during overground walking, influencing kinetic and kinematic gait parameters. An alternative is using an instrumented treadmill that benefits from capturing many steady-state gait cycles at a constant speed quickly. The goal of this study was (1) to compare overground with treadmill locomotion and (2) to compare the instrumented treadmill with force plates for dogs kinetics measurements.Twelve client-owned dogs were measured during treadmill trotting while the treadmill was placed on force plates. Additionally, the dogs were measured during trotting along an alley over eight force plates. Bland-Altman plots, Pearson's (r), and concordance correlation coefficients (rc) were computed to explore the relative and general agreement between the measurement methods and overground and treadmill trotting.Overground and treadmill trotting gave an excellent agreement in peak vertical forces and impulses (r > 0.9, rc > 0.9). The instrumented treadmill showed similar force-time curves in shape and size and provided an excellent congruity for all parameters compared to force plates (r > 0.8, rc > 0.8).As a reliable tool in measuring key gait parameters, the instrumented treadmill may benefit from fast and reproducible data comparable to overground trotting.

Greater Breast Support Alters Trunk and Knee Joint Biomechanics Commonly Associated With Anterior Cruciate Ligament Injury.

ObjectiveThe female breast is a passive tissue with little intrinsic support. Therefore, women rely on external breast support (sports bras) to control breast motion during athletic tasks. Research has demonstrated that lower levels of breast support are associated with altered trunk and pelvis movement patterns during running, a common athletic task. However, no previous study has identified the effect of sports bra support on movement patterns during other athletic tasks including landing. Therefore, the purpose of this study was to examine the effects of breast support on trunk and knee joint biomechanics in female collegiate athletes during a double-leg landing task.MethodsFourteen female collegiate athletes completed five double-leg landing trials in each of three different sports bra conditions: no support, low support, and high support. A 10-camera motion capture system (250 Hz, Qualisys, Goteburg, Sweden) and two force platforms (1,250 Hz, AMTI, Watertown, MA, USA) were used to collect three-dimensional kinematics and ground reaction forces simultaneously. Visual 3D was used to calculate trunk segment and knee joint angles and moments. Custom software (MATLAB 2021a) was used to determine discrete values of dependent variables including vertical breast displacement, knee joint and trunk segment angles at initial contact and 100 ms post-initial contact, and peak knee joint moments. A repeated measures analysis of covariance with post-hoc paired samples t-tests were used to evaluate the effect of breast support on landing biomechanics.ResultsIncreasing levels of breast support were associated with reductions in peak knee flexion (Right: p = 0.008; Left: p = 0.029) and peak knee valgus angles (Right: p = 0.011; Left: p = 0.003) as well as reductions in peak knee valgus moments (Right: p = 0.033; Left: p = 0.013). There were no changes in peak knee extension moments (Right: p = 0.216; Left: p = 0.261). Increasing levels of breast support were associated with greater trunk flexion angles at initial contact (p = 0.024) and greater peak trunk flexion angles (p = 0.002).ConclusionsLower levels of breast support are associated with knee joint and trunk biomechanical profiles suggested to increase ACL injury risk.

Adaptations for bipedal walking: Musculoskeletal structure and three-dimensional joint mechanics of humans and bipedal chimpanzees (Pan troglodytes)

Humans are unique among apes and other primates in the musculoskeletal design of their lower back, pelvis, and lower limbs. Here, we describe the three-dimensional ground reaction forces and lower/hindlimb joint mechanics of human and bipedal chimpanzees walking over a full stride and test whether: 1) the estimated limb joint work and power during the stance phase, especially the single-support period, is lower in humans than bipedal chimpanzees, 2) the limb joint work and power required for limb swing is lower in humans than in bipedal chimpanzees, and 3) the estimated total mechanical power during walking, accounting for the storage of passive elastic strain energy in humans, is lower in humans than in bipedal chimpanzees. Humans and bipedal chimpanzees were compared at matched dimensionless and dimensional velocities. Our results indicate that humans walk with significantly less work and power output in the first double-support period and the single-support period of stance, but markedly exceed chimpanzees in the second double-support period (i.e., push-off). Humans generate less work and power in limb swing, although the species difference in limb swing power was not statistically significant. We estimated that total mechanical positive ‘muscle fiber’ work and power were 46.9% and 35.8% lower, respectively, in humans than in bipedal chimpanzees at matched dimensionless speeds. This is due in part to mechanisms for the storage and release of elastic energy at the ankle and hip in humans. Furthermore, these results indicate distinct ‘heel strike’ and ‘lateral balance’ mechanics in humans and bipedal chimpanzees and suggest a greater dissipation of mechanical energy through soft tissue deformations in humans. Together, our results document important differences between human and bipedal chimpanzee walking mechanics over a full stride, permitting a more comprehensive understanding of the mechanics and energetics of chimpanzee bipedalism and the evolution of hominin walking.

Leveraging explainable machine learning to identify gait biomechanical parameters associated with anterior cruciate ligament injury

Anterior cruciate ligament (ACL) deficient and reconstructed knees display altered biomechanics during gait. Identifying significant gait changes is important for understanding normal and ACL function and is typically performed by statistical approaches. This paper focuses on the development of an explainable machine learning (ML) empowered methodology to: (i) identify important gait kinematic, kinetic parameters and quantify their contribution in the diagnosis of ACL injury and (ii) investigate the differences in sagittal plane kinematics and kinetics of the gait cycle between ACL deficient, ACL reconstructed and healthy individuals. For this aim, an extensive experimental setup was designed in which three-dimensional ground reaction forces and sagittal plane kinematic as well as kinetic parameters were collected from 151 subjects. The effectiveness of the proposed methodology was evaluated using a comparative analysis with eight well-known classifiers. Support Vector Machines were proved to be the best performing model (accuracy of 94.95%) on a group of 21 selected biomechanical parameters. Neural Networks accomplished the second best performance (92.89%). A state-of-the-art explainability analysis based on SHapley Additive exPlanations (SHAP) and conventional statistical analysis were then employed to quantify the contribution of the input biomechanical parameters in the diagnosis of ACL injury. Features, that would have been neglected by the traditional statistical analysis, were identified as contributing parameters having significant impact on the ML model’s output for ACL injury during gait.

Discriminating features of ground reaction forces in overweight old and young adults during walking using functional principal component analysis

BackgroundLimited attention has been paid to age- or body size-related changes in the ground reaction forces (GRF) during walking despite their strong associations with lower limb injuries and pathology. Research questionDo the features of GRF during walking associate with age or body size? MethodsFifty-four participants were subdivided into four groups according to their age and body size: overweight old (n = 12), non-overweight old (n = 13), overweight young (n = 13), and non-overweight young (n = 16). Participants were asked to walk at their self-selected speeds on level ground with force plates embedded in the center of walkway. Functional principal component analysis (FPCA) was performed to extract major modes of variation and functional principal component scores (FPCs) in three-dimensional GRFs. Analysis of variance models were employed to investigate the effect of age, body size, or their interactions on the FPCs of each component of the GRF, with the adjustment to gait speed. ResultsSignificant age and body size effects were observed in FPC1 across all three-dimensional GRF. Both overweight and older groups showed greater braking force after heel-strike and greater propulsive forces during pre-swing when compared to the non-overweight and younger groups, respectively. The overweight old group displayed greater medial forces during mid-stance and the overweight young group showed prominently larger medial forces during pre-swing, while non-overweight old showed a tendency of flatter medial-lateral GRF waveforms during the entire stance phase. FPC2 revealed that only body size had an effect on three-dimensional GRF with the highest FPC2 scores in the overweight old group. SignificanceThree-dimensional GRF during walking could be altered by the body size and age, which were more pronounced in the overweight and older group. The more dynamic GRF pattern with greater and/or lower peaks could be contributing factors to the increased joint load and injury rates observed in overweight aged individuals.