A Review of CT-Based Fracture Risk Assessment with Finite Element Modeling and Machine Learning.

PURPOSE OF REVIEW: We reviewed advances over the past 3 years in assessment of fracture risk based on CT scans, considering methods that use finite element models, machine learning, or a combination of both. RECENT FINDINGS: Several studies have demonstrated that CT-based assessment of fracture risk, using finite element modeling or biomarkers derived from machine learning, is equivalent to currently used clinical tools. Phantomless calibration of CT scans for bone mineral density enables accurate measurements from routinely taken scans. This opportunistic use of CT scans for fracture risk assessment is facilitated by high-quality automated segmentation with deep learning, enabling workflows that do not require user intervention. Modeling of more realistic and diverse loading conditions, as well as improved modeling of fracture mechanisms, has shown promise to enhance our understanding of fracture processes and improve the assessment of fracture risk beyond the performance of current clinical tools. CT-based screening for fracture risk is effective and, by analyzing scans that were taken for other indications, could be used to expand the pool of people screened, therefore improving fracture prevention. Finite element modeling and machine learning both provide valuable tools for fracture risk assessment. Future approaches should focus on including more loading-related aspects of fracture risk.

Geometric determinants of the mechanical behavior of image-based finite element models of the intervertebral disc.

The intervertebral disc is an important structure for load transfer through the spine. Its injury and degeneration have been linked to pain and spinal fractures. Disc injury and spine fractures are associated with high stresses; however, these stresses cannot be measured, necessitating the use of finite element (FE) models. These models should include the disc's complex structure, as changes in disc geometry have been linked to altered mechanical behavior. However, image-based models using disc-specific structures have yet to be established. This study describes a multiphasic FE modeling approach for noninvasive estimates of subject-specific intervertebral disc mechanical behavior based on medical imaging. The models (n = 22) were used to study the influence of disc geometry on the predicted global mechanical response (moments and forces), internal local disc stresses, and tractions at the interface between the disc and the bone. Disc geometry was found to have a strong influence on the predicted moments and forces on the disc (R2 = 0.69-0.93), while assumptions regarding the side curvature (bulge) of the disc had only a minor effect. Strong variability in the predicted internal disc stresses and tractions was observed between the models (mean absolute differences of 5.1%-27.7%). Disc height had a systematic influence on the internal disc stresses and tractions at the disc-to-bone interface. The influence of disc geometry on mechanics highlights the importance of disc-specific modeling to estimate disc injury risk, loading on the adjacent vertebral bodies, and the mechanical environment present in disc tissues.

How accurately do finite element models predict the fall impact response of ex vivo specimens augmented by prophylactic intramedullary nailing?

Hip fracture prevention approaches like prophylactic augmentation devices have been proposed to strengthen the femur and prevent hip fracture in a fall scenario. The aim of this study was to validate the finite element model (FEM) of specimens augmented by prophylactic intramedullary nailing in a simulated sideways fall impact against ex vivo experimental data. A dynamic inertia-driven sideways fall simulator was used to test six cadaveric specimens (3 females, 3 males, age 63-83 years) prophylactically implanted with an intramedullary nailing system used to augment the femur. Impact force measurements, pelvic deformation, effective pelvic stiffness, and fracture outcomes were compared between the ex vivo experiments and the FEMs. The FEMs over-predicted the effective pelvic stiffness for most specimens and showed variability in terms of under- and over-predicting peak impact force and pelvis compression depending on the specimen. A significant correlation was found for time to peak impact force when comparing ex vivo and FEM data. No femoral fractures were found in the ex vivo experiments, but two specimens sustained pelvic fractures. These two pelvis fractures were correctly identified by the FEMs, but the FEMs made three additional false-positive fracture identifications. These validation results highlight current limitations of these sideways fall impact models specific to the inclusion of an orthopaedic implant. These FEMs present a conservative strategy for fracture prediction in future applications. Further evaluation of the modelling approaches used for the bone-implant interface is recommended for modelling augmented specimens, alongside the importance of maintaining well-controlled experimental conditions.

The efficacy of femoral augmentation for hip fracture prevention using ceramic-based cements: A preliminary experimentally-driven finite element investigation.

Femoral fractures due to sideways falls continue to be a major cause of concern for the elderly. Existing approaches for the prevention of these injuries have limited efficacy. Prophylactic femoral augmentation systems, particularly those involving the injection of ceramic-based bone cements, are gaining more attention as a potential alternative preventative approach. We evaluated the mechanical effectiveness of three variations of a bone cement injection pattern (basic ellipsoid, hollow ellipsoid, small ellipsoid) utilizing finite element simulations of sideways fall impacts. The basic augmentation pattern was tested with both high- and low-strength ceramic-based cements. The cement patterns were added to the finite element models (FEMs) of five cadaveric femurs, which were then subject to simulated sideways falls at seven impact velocities ranging from 1.0 m/s to 4.0 m/s. Peak impact forces and peak acetabular forces were examined, and failure was evaluated using a strain-based criterion. We found that the basic HA ellipsoid provided the highest increases in both the force at the acetabulum of the impacted femur ("acetabular force", 55.0% ± 22.0%) and at the force plate ("impact force", 37.4% ± 15.8%). Changing the cement to a weaker material, brushite, resulted in reduced strengthening of the femur (45.2% ± 19.4% acetabular and 30.4% ± 13.0% impact). Using a hollow version of the ellipsoid appeared to have no effect on the fracture outcome and only a minor effect on the other metrics (54.1% ± 22.3% acetabular force increase and 35.3% ± 16.0% impact force increase). However, when the outer two layers of the ellipsoid were removed (small ellipsoid), the force increases that were achieved were only 9.8% ± 5.5% acetabular force and 8.2% ± 4.1% impact force. These results demonstrate the importance of supporting the femoral neck cortex to prevent femoral fractures in a sideways fall, and provide plausible options for prophylactic femoral augmentation. As this is a preliminary study, the surgical technique, the possible effects of trabecular bone damage during the augmentation process, and the effect on the blood supply to the femoral head must be assessed further.

The Influence of Fall Direction and Hip Protector on Fracture Risk: FE Model Predictions Driven by Experimental Data.

Hip fractures in older adults, which often lead to lasting impairments and an increased risk of mortality, are a major public health concern. Hip fracture risk is multi-factorial, affected by the risk of falling, the load acting on the femur, and the load the femur can withstand. This study investigates the influence of impact direction on hip fracture risk and hip protector efficacy. We simulated falls for 4 subjects, in 7 different impact directions (15° and 30° anterior, lateral, and 15°, 30°, 60°, and 90° posterior) at two different impact velocities (2.1 and 3.1 m/s), all with and without hip protector, using previously validated biofidelic finite element models. We found the highest number of fractures and highest fragility ratios in lateral and 15° posterior impacts. The hip protector attenuated femur forces by 23-49 % for slim subjects under impact directions that resulted in fractures (30° anterior to 30° posterior). The hip protector prevented all fractures (6/6) for 2.1 m/s impacts, but only 10% of fractures for 3.1 m/s impacts. Our results provide evidence that, regarding hip fracture risk, posterior-lateral impacts are as dangerous as lateral impacts, and they support the efficacy of soft-shell hip protectors for anterior- and posterior-lateral impacts.

Prophylactic augmentation implants in the proximal femur for hip fracture prevention: An in silico investigation of simulated sideways fall impacts.

Femoral fractures from sideways falls in the elderly are associated with significant rates of morbidity and mortality. Approaches to prevent these catastrophic injuries include pharmacological treatments, which have limited efficacy. Prophylactic femoral augmentation systems are a promising alternative that are gaining prominence by addressing the most debilitating osteoporosis-related fracture. We have developed finite element models (FEMs) of a novel experimental sideways fall simulator for cadavers. By virtue of the range of specimens and injury outcomes, these FEMs are well-suited to the evaluation of such implants. The purpose of this study was to use the FEMs to evaluate the mechanical effectiveness of three different prophylactic femoral augmentation systems. Models of the Y-Strut® (Hyprevention®, Pessac, France), Gamma Nail® (Stryker, Kalamazoo, USA), and a simple lag screw femoral fracture implant systems were placed into FEMs of five cadaveric pelvis-femur constructs embedded in a soft tissue surrogate, which were then subject to simulated sideways falls at seven impact velocities. Femur-only FEMs were also evaluated. Peak impact forces and peak acetabular forces were examined, and failure was evaluated using a strain-based criterion. We found that the femoral augmentation systems increased the peak forces prior to fracture, but were unable to prevent fracture for severe impacts. The Gamma Nail® system consistently produced the largest strength increases relative to the unaugmented femur for all five specimens in both the pendulum-drop FEMs and the femur-only simulations. In some cases, the same implant appeared to cause fractures in the acetabulum. The femur-only FEMs showed larger force increases than the pendulum-drop simulations, which suggests that the results of the femur-only simulations may not represent sideways falls as accurately as the soft tissue-embedded pendulum-drop simulations. The results from this study demonstrate the ability to simulate a high energy phenomenon and the effect of implants in an in silico environment. The results also suggest that implants could increase the force applied to the proximal femur during impact. Fracture outcomes from the tested implants can be used to inform the design of future devices, which reaffirms the value of modelling with biofidelic considerations in the implant design process. To the authors' knowledge, this is the first paper to use more complex biofidelic FEMs to assess prophylactic femoral augmentation methods.

Finite element derived femoral strength is a better predictor of hip fracture risk than aBMD in the AGES Reykjavik study cohort.

Hip fractures associated with a high economic burden, loss of independence, and a high rate of post-fracture mortality, are a major health concern for modern societies. Areal bone mineral density is the current clinical metric of choice when assessing an individual's future risk of fracture. However, this metric has been shown to lack sensitivity and specificity in the targeted selection of individuals for preventive interventions. Although femoral strength derived from computed tomography based finite element models has been proposed as an alternative based on its superior femoral strength prediction ex vivo, such predictions have only shown marginal or no improvement for assessing hip fracture risk. This study compares finite element derived femoral strength to aBMD as a metric for hip fracture risk assessment in subjects (N = 601) from the AGES Reykjavik Study cohort and analyses the dependence of femoral strength predictions and classification accuracy on the material model and femoral loading alignment. We found hip fracture classification based on finite element derived femoral strength to be significantly improved compared to aBMD. Finite element models with non-linear material models performed better at classifying hip fractures compared to finite element models with linear material models and loading alignments with low internal rotation and adduction, which do not correspond to weak femur alignments, were found to be most suitable for hip fracture classification.

Validation of 3D finite element models from simulated DXA images for biofidelic simulations of sideways fall impact to the hip.

Computed tomography (CT)-derived finite element (FE) models have been proposed as a tool to improve the current clinical assessment of osteoporosis and personalized hip fracture risk by providing an accurate estimate of femoral strength. However, this solution has two main drawbacks, namely: (i) 3D CT images are needed, whereas 2D dual-energy x-ray absorptiometry (DXA) images are more generally available, and (ii) quasi-static femoral strength is predicted as a surrogate for fracture risk, instead of predicting whether a fall would result in a fracture or not. The aim of this study was to combine a biofidelic fall simulation technique, based on 3D computed tomography (CT) data with an algorithm that reconstructs 3D femoral shape and BMD distribution from a 2D DXA image. This approach was evaluated on 11 pelvis-femur constructs for which CT scans, ex vivo sideways fall impact experiments and CT-derived biofidelic FE models were available. Simulated DXA images were used to reconstruct the 3D shape and bone mineral density (BMD) distribution of the left femurs by registering a projection of a statistical shape and appearance model with a genetic optimization algorithm. The 2D-to-3D reconstructed femurs were meshed, and the resulting FE models inserted into a biofidelic FE modeling pipeline for simulating a sideways fall. The median 2D-to-3D reconstruction error was 1.02 mm for the shape and 0.06 g/cm3 for BMD for the 11 specimens. FE models derived from simulated DXAs predicted the outcome of the falls in terms of fracture versus non-fracture with the same accuracy as the CT-derived FE models. This study represents a milestone towards improved assessment of hip fracture risk based on widely available clinical DXA images.

Empirical relationships between bone density and ultimate strength: A literature review.

INTRODUCTION: Ultimate strength-density relationships for bone have been reported with widely varying results. Reliable bone strength predictions are crucial for many applications that aim to assess bone failure. Bone density and bone morphology have been proposed to explain most of the variance in measured bone strength. If this holds true, it could lead to the derivation of a single ultimate strength-density-morphology relationship for all anatomical sites. METHODS: All relevant literature was reviewed. Ultimate strength-density relationships derived from mechanical testing of human bone tissue were included. The reported relationships were translated to ultimate strength-apparent density relationships and normalized with respect to strain rate. Results were grouped based on bone tissue type (cancellous or cortical), anatomical site, and loading mode (tension vs. compression). When possible, the relationships were compared to existing ultimate strength-density-morphology relationships. RESULTS: Relationships that considered bone density and morphology covered the full spectrum of eight-fold inter-study difference in reported compressive ultimate strength-density relationships for trabecular bone. This was true for studies that tested specimens in different loading direction and tissue from different anatomical sites. Sparse data was found for ultimate strength-density relationships in tension and for cortical bone properties transverse to the main loading axis of the bone. CONCLUSIONS: Ultimate strength-density-morphology relationships could explain measured strength across anatomical sites and loading directions. We recommend testing of bone specimens in other directions than along the main trabecular alignment and to include bone morphology in studies that investigate bone material properties. The lack of tensile strength data did not allow for drawing conclusions on ultimate strength-density-morphology relationships. Further studies are needed. Ideally, these studies would investigate both tensile and compressive strength-density relationships, including morphology, to close this gap and lead to more accurate evaluation of bone failure.

Implicit and explicit finite element models predict the mechanical response of calcium phosphate-titanium cranial implants.

The structural integrity of cranial implants is of great clinical importance, as they aim to provide cerebral protection after neurosurgery or trauma. With the increased use of patient-specific implants, the mechanical response of each implant cannot be characterized experimentally in a practical way. However, computational models provide an excellent possibility for efficiently predicting the mechanical response of patient-specific implants. This study developed finite element models (FEMs) of titanium-reinforced calcium phosphate (CaP-Ti) implants. The models were validated with previously obtained experimental data for two different CaP-Ti implant designs (D1 and D2), in which generically shaped implant specimens were loaded in compression at either quasi-static (1 mm/min) or impact (5 kg, 1.52 m/s) loading rates. The FEMs showed agreement with experimental data in the force-displacement response for both implant designs. The implicit FEMs predicted the peak load with an underestimation for D1 (9%) and an overestimation for D2 (11%). Furthermore, the shape of the force-displacement curves were well predicted. In the explicit FEMs, the first part of the force-displacement response showed 5% difference for D1 and 2% difference for D2, with respect to the experimentally derived peak loads. The explicit FEMs efficiently predicted the maximum displacements with 1% and 4% difference for D1 and D2, respectively. Compared to the CaP-Ti implant, an average parietal cranial bone FEM showed a stiffer response, greater energy absorption and less deformation under the same impact conditions. The framework developed for modelling the CaP-Ti implants has a potential for modelling CaP materials in other composite implants in future studies since it only used literature based input and matched boundary conditions. Furthermore, the developed FEMs make an important contribution to future evaluations of patient-specific CaP-Ti cranial implant designs in various loading scenarios.

Explicit Finite Element Models Accurately Predict Subject-Specific and Velocity-Dependent Kinetics of Sideways Fall Impact.

The majority of hip fractures in the elderly are the result of a fall from standing or from a lower height. Current injury models focus mostly on femur strength while neglecting subject-specific loading. This article presents an injury modeling strategy for hip fractures related to sideways falls that takes subject-specific impact loading into account. Finite element models (FEMs) of the human body were used to predict the experienced load and the femoral strength in a single model. We validated these models for their predicted peak force, effective pelvic stiffness, and fracture status against matching ex vivo sideways fall impacts (n = 11) with a trochanter velocity of 3.1 m/s. Furthermore, they were compared to sideways impacts of volunteers with lower impact velocities that were previously conducted by other groups. Good agreement was found between the ex vivo experiments and the FEMs with respect to peak force (root mean square error [RMSE] = 10.7%, R2 = 0.85) and effective pelvic stiffness (R2 = 0.92, RMSE = 12.9%). The FEMs were predictive of the fracture status for 10 out of 11 specimens. Compared to the volunteer experiments from low height, the FEMs overestimated the peak force by 25% for low BMI subjects and 8% for high BMI subjects. The effective pelvic stiffness values that were derived from the FEMs were comparable to those derived from impacts with volunteers. The force attenuation from the impact surface to the femur ranged between 27% and 54% and was highly dependent on soft tissue thickness (R2 = 0.86). The energy balance in the FEMS showed that at the time of peak force 79% to 93% of the total energy is either kinetic or was transformed to soft tissue deformation. The presented FEMs allow for direct discrimination between fracture and nonfracture outcome for sideways falls and bridge the gap between impact testing with volunteers and impact conditions representative of real life falls. © 2019 American Society for Bone and Mineral Research.

Subject-specific ex vivo simulations for hip fracture risk assessment in sideways falls.

The risk of hip fracture of a patient due to a fall can be described from a mechanical perspective as the capacity of the femur to withstand the force that it experiences in the event of a fall. So far, impact forces acting on the lateral aspect of the pelvic region and femur strength have been investigated separately. This study used inertia-driven cadaveric impact experiments that mimic falls to the side from standing in order to evaluate the subject-specific force applied to the hip during impact and the fracture outcome in the same experimental model. Eleven fresh-frozen pelvis-femur constructs (6 female, 5 male, age = 77 years (SD = 13 years), BMI = 22.8 kg/m2 (SD = 7.8 kg/m2), total hip aBMD = 0.734 g/cm2 (SD = 0.149 g/cm2)), were embedded into soft tissue surrogate material that matched subject-specific mass and body shape. The specimens were attached to metallic lower-limb constructions with subject-specific masses and subjected to an inverted pendulum motion. Impact forces were recorded with a 6-axis force plate at 10,000 Hz and three dimensional deflections in the pelvic region were tracked with two high-speed cameras at 5000 Hz. Of the 11 specimens, 5 femur fractures and 3 pelvis fractures were observed. Three specimens did not fracture. aBMD alone did not reliably separate femur fractures from non-fractures. However, a mechanical risk ratio, which was calculated as the impact force divided by aBMD, classified specimens reliably into femur fractures and non-fractures. Single degree of freedom models, based on specimen kinetics, were able to predict subject-specific peak impact forces (RMSE = 2.55% for non-fractures). This study provides direct evidence relating subject-specific impact forces and subject-specific strength estimates and improves the assessment of the mechanical risk of hip fracture for a specific femur/pelvis combination in a sideways fall.

On the internal reaction forces, energy absorption, and fracture in the hip during simulated sideways fall impact.

The majority of hip fractures have been reported to occur as a result of a fall with impact to the greater trochanter of the femur. Recently, we developed a novel cadaveric pendulum-based hip impact model and tested two cadaveric femur-pelvis constructs, embedded in a soft tissue surrogate. The outcome was a femoral neck fracture in a male specimen while a female specimen had no fracture. The aim of the present study was, first, to develop a methodology for constructing and assessing the accuracy of explicit Finite Element Models (FEMs) for simulation of sideways falls to the hip based on the experimental model. Second, to use the FEMs for quantifying the internal reaction forces and energy absorption in the hip during impact. Third, to assess the potential of the FEMs in terms of separating a femoral fracture endpoint from a non-fracture endpoint. Using a non-linear, strain rate dependent, and heterogeneous material mapping strategy for bone tissue in these models, we found the FEM-derived results to closely match the experimental test results in terms of impact forces and displacements of pelvic video markers up to the time of peak impact force with errors below 10%. We found the internal reaction forces in the femoral neck on the impact side to be approximately 35% lower than the impact force measured between soft tissue and ground for both specimens. In addition, we found the soft tissue to be the component that absorbed the largest part of the energy of the tissue types in the hip region. Finally, we found surface strain patterns derived from FEM results to match the fracture location and extent based on post testing x-rays of the specimens. This is the first study with quantitative data on the energy absorption in the pelvic region during a sideways fall.

Correction: On the internal reaction forces, energy absorption, and fracture in the hip during simulated sideways fall impact.

[This corrects the article DOI: 10.1371/journal.pone.0200952.].

A novel sideways fall simulator to study hip fractures ex vivo.

Falls to the side are the leading cause of hip fractures in the elderly. The load that a person experiences during a fall cannot be measured with volunteers for ethical reasons. To evaluate injurious loads, while considering relevant energy input and body posture for a sideways fall, a subject-specific cadaveric impact experiment was developed. Full cadaveric femur-pelvis constructs (N = 2) were embedded in surrogate soft tissue material and attached to metallic surrogate lower limbs. The specimens were then subjected to an inverted pendulum motion, simulating a fall to the side with an impact to the greater trochanter. The load at the ground and the deformation of the pelvis were evaluated using a 6-axis force transducer and two high-speed cameras. Post-test, a trauma surgeon (PG) evaluated specimen injuries. Peak ground contact forces were 7132 N and 5641 N for the fractured and non-fractured specimen, respectively. We observed a cervical fracture of the femur in one specimen and no injuries in a second specimen, showing that the developed protocol can be used to differentiate between specimens at high and low fracture risk.