Use of subject-specific models to detect fatigue-related changes in running biomechanics: a random forest approach

Running biomechanics are affected by fatiguing or prolonged runs. However, no evidence to date has conclusively linked this effect to running-related injury (RRI) development or performance implications. Previous investigations using subject-specific models in running have demonstrated higher accuracy than group-based models, however, this has been infrequently applied to fatigue. In this study, two experiments were conducted to determine whether subject-specific models outperformed group-based models to classify running biomechanics during non-fatigued and fatigued conditions. In the first experiment, 16 participants performed four treadmill runs at or around the maximal lactate steady state. In the second experiment, nine participants performed five prolonged runs using commercial wearable devices. For each experiment, two segments were extracted from each trial from early and late in the run. For each participant, a random forest model was applied with a leave-one-run-out cross-validation to classify between the early (non-fatigued) and late (fatigued) segments. Additionally, group-based classifiers with a leave-one-subject-out cross validation were constructed. For experiment 1, mean classification accuracies for the single-subject and group-based classifiers were 68.2 ± 8.2% and 57.0 ± 8.9%, respectively. For experiment 2, mean classification accuracies for the single-subject and group-based classifiers were 68.9 ± 17.1% and 61.5 ± 11.7%, respectively. Variable importance rankings were consistent within participants, but these rankings differed from each participant to those of the group. Although the classification accuracies were relatively low, these findings highlight the advantage of subject-specific classifiers to detect changes in running biomechanics with fatigue and indicate the potential of using big data and wearable technology approaches in future research to determine possible connections between biomechanics and RRI.

A Morphologically Individualized Deep Learning Brain Injury Model

The brain injury modeling community has recommended improving model subject specificity and simulation efficiency. Here, we extend an instantaneous (< 1 sec) convolutional neural network (CNN) brain model based on the anisotropic Worcester Head Injury Model (WHIM) V1.0 to account for strain differences due to individual morphological variations. Linear scaling factors relative to the generic WHIM along the three anatomical axes are used as additional CNN inputs. To generate training samples, the WHIM is randomly scaled to pair with augmented head impacts randomly generated from real-world data for simulation. An estimation of voxelized peak maximum principal strain of the whole-brain is said to be successful when the linear regression slope and Pearson's correlation coefficient relative to directly simulated do not deviate from 1.0 (when identical) by more than 0.1. Despite a modest training dataset (N = 1363 vs. ∼5.7 k previously), the individualized CNN achieves a success rate of 86.2% in cross-validation for scaled model responses, and 92.1% for independent generic model testing for impacts considered as complete capture of kinematic events. Using 11 scaled subject-specific models (with scaling factors determined from pre-established regression models based on head dimensions and sex and age information, and notably, without neuroimages), the morphologically individualized CNN remains accurate for impacts that also yield successful estimations for the generic WHIM. The individualized CNN instantly estimates subject-specific and spatially detailed peak strains of the entire brain and thus, supersedes others that report a scalar peak strain value incapable of informing the location of occurrence. This tool could be especially useful for youths and females due to their anticipated greater morphological differences relative to the generic model, even without the need for individual neuroimages. It has potential for a wide range of applications for injury mitigation purposes and the design of head protective gears. The voxelized strains also allow for convenient data sharing and promote collaboration among research groups.

Calibration of Holzapfel-Gasser-Ogden collateral ligament properties in a hybrid post-arthroplasty knee joint model for laxity testing

Knee collateral ligaments play a vital role in providing frontal-plane stability in post-total knee arthroplasty (TKA) knees. Finite element models can utilize computationally efficient one-dimensional springs or more physiologically accurate three-dimensional continuum elements like the Holzapfel-Gasser-Ogden (HGO) formulation. However, there is limited literature defining subject-specific mechanical properties, particularly for the HGO model. In this study, we propose a co-simulation framework to obtain subject-specific material parameters for an HGO-based finite element ligament model integrated into a rigid-body model of the post-TKA knee. Our approach achieves comparable accuracy to spring formulations while significantly reducing coefficient calibration time and demonstrating improved correlation with reference knee kinematics and ligament strains throughout the tested loading range.

Comparison of small female occupant model responses with experimental data in a reclined posture

Objective: The objective of the current study was to compare the GHBMC female model responses with in-house sled test data for three small female post mortem human surrogates (PMHS) at 32 km/h and a seatback recline angle of 45 degrees. The kinematics and the seatbelt forces were used to compare the female PMHS and model responses. The study aimed to identify updates that may be needed to the model.Methods: In-house experimental sled test kinematic and seatbelt response data for the small females were obtained. The 5^th female GHBMC was simulated with the same boundary conditions as in the experiments. In addition, using the PMHS computed tomography (CT) and test environment scans, the female model geometry was updated to a subject-specific model for one of the specimens, and the models were simulated to obtain 5^th female and subject-specific model responses. The kinematic response and the seatbelt forces for the two models were compared with the average of the three experimental data.Results: The head, T8 and L4 excursions, head and pelvis accelerations and seatbelt forces for the two female models were compared with the experimental data. The model responses were in agreement with the PMHS; however, the subject-specific model showed a closer agreement with the kinematic response. The subject-specific model did not submarine as in the experiments, whereas the 5^th female model submarined. However, the subject-specific model showed 20% higher seatbelt forces than the PMHS.Conclusion: This study showed that anthropometric differences may significantly alter occupant kinematics in reclined posture and need to be incorporated to investigate kinematics and injury mechanisms. The next step of the study involves incorporating age-specific material changes and investigating the subject-specific injury mechanisms. The results will be useful to develop countermeasures for autonomous vehicles.

Methodology to geometrically age human body models to average and subject-specific anthropometrics, demonstrated using a small stature female model assessed in a side impact

The aged population has been associated with an increased risk of injury in car-crash, creating a critical need for improved assessment of safety systems. Finite element human body models (HBMs) have been proposed, but require representative geometry of the aged population and high mesh quality. A new hybrid Morphing-CAD methodology was applied to a 26-year-old (YO) 5^th percentile female model to create average 75YO and subject-specific 86YO HBMs. The method achieved accurate morphing targets while retaining high mesh quality. The three HBMs were integrated into a side sled impact test demonstrating similar kinematic response but differing rib fracture patterns.

Uncertainty quantification in subject-specific estimation of local vessel mechanical properties

Quantitative estimation of local mechanical properties remains critically important in the ongoing effort to elucidate how blood vessels establish, maintain, or lose mechanical homeostasis. Recent advances based on panoramic digital image correlation (pDIC) have made high-fidelity 3D reconstructions of small-animal (e.g., murine) vessels possible when imaged in a variety of quasi-statically loaded configurations. While we have previously developed and validated inverse modeling approaches to translate pDIC-measured surface deformations into biomechanical metrics of interest, our workflow did not heretofore include a methodology to quantify uncertainties associated with local point estimates of mechanical properties. This limitation has compromised our ability to infer biomechanical properties on a subject-specific basis, such as whether stiffness differs significantly between multiple material locations on the same vessel or whether stiffness differs significantly between multiple vessels at a corresponding material location. In the present study, we have integrated a novel uncertainty quantification and propagation pipeline within our inverse modeling approach, relying on empirical and analytic Bayesian techniques. To demonstrate the approach, we present illustrative results for the ascending thoracic aorta from three mouse models, quantifying uncertainties in constitutive model parameters as well as circumferential and axial tangent stiffness. Our extended workflow not only allows parameter uncertainties to be systematically reported, but also facilitates both subject-specific and group-level statistical analyses of the mechanics of the vessel wall.

A generic musculoskeletal model of the juvenile lower limb for biomechanical analyses of gait

The aim of this study was to develop a generic musculoskeletal model of a healthy 10-year-old child and examine the effects of geometric scaling on the calculated values of lower-limb muscle forces during gait. Subject-specific musculoskeletal models of five healthy children were developed from in vivo MRI data, and these models were subsequently used to create a generic juvenile (GJ) model. Calculations of lower-limb muscle forces for normal walking obtained from two scaled-generic versions of the juvenile model (SGJ1 and SGJ2) were evaluated against corresponding results derived from an MRI-based model of one subject (SSJ1). The SGJ1 and SGJ2 models were created by scaling the GJ model using gait marker positions and joint centre locations derived from MRI imaging, respectively. Differences in the calculated values of peak isometric muscle forces and muscle moment arms between the scaled-generic models and MRI-based model were relatively small. Peak isometric muscle forces calculated for SGJ1 and SGJ2 were respectively 2.2% and 3.5% lower than those obtained for SSJ1. Model-predicted muscle forces for SGJ2 agreed more closely with calculations obtained from SSJ1 than corresponding results derived from SGJ1. These results suggest that accurate estimates of muscle forces during gait may be obtained by scaling generic juvenile models based on joint centre locations. The generic juvenile model developed in this study may be used as a template for creating subject-specific musculoskeletal models of normally-developing children in studies aimed at describing lower-limb muscle function during gait.

A reduced-order flow model for vocal fold vibration: from idealized to subject-specific models

We present a reduced-order model for fluid-structure interaction (FSI) simulation of vocal fold vibration during phonation. This model couples the three-dimensional (3D) tissue mechanics and a one-dimensional (1D) flow model that is derived from the momentum and mass conservation equations for the glottal airflow. The effects of glottal entrance and pressure loss in the glottis are incorporated in the flow model. We consider both idealized vocal fold geometries and subject-specific anatomical geometries segmented from the MRI images of rabbits. For the idealized vocal fold geometries, we compare the simulation results from the 1D/3D hybrid FSI model with those from the full 3D FSI simulation based on an immersed-boundary method. For the subject-specific geometries, we incorporate previously estimated tissue properties for individual samples and compare the results with those from the high-speed imaging experiment of in vivo phonation. In both setups, the comparison shows good agreement in the vibration frequency, amplitude, phase delay, and deformation pattern of the vocal fold, which suggests potential application of the present approach for future patient-specific modeling.

A Lumped-Parameter Subject-Specific Model of Blood Volume Response to Fluid Infusion

This paper presents a lumped-parameter model that can reproduce blood volume response to fluid infusion. The model represents the fluid shift between the intravascular and interstitial compartments as the output of a hypothetical feedback controller that regulates the ratio between the volume changes in the intravascular and interstitial fluid at a target value (called "target volume ratio"). The model is characterized by only three parameters: the target volume ratio, feedback gain (specifying the speed of fluid shift), and initial blood volume. This model can obviate the need to incorporate complex mechanisms involved in the fluid shift in reproducing blood volume response to fluid infusion. The ability of the model to reproduce real-world blood volume response to fluid infusion was evaluated by fitting it to a series of data reported in the literature. The model reproduced the data accurately with average error and root-mean-squared error (RMSE) of 0.6 and 9.5% across crystalloid and colloid fluids when normalized by the underlying responses. Further, the parameters derived for the model showed physiologically plausible behaviors. It was concluded that this simple model may accurately reproduce a variety of blood volume responses to fluid infusion throughout different physiological states by fitting three parameters to a given dataset. This offers a tool that can quantify the fluid shift in a dataset given the measured fractional blood volumes.