Investigating the Influence of Limited CT Scan Coverage Using an Automated Workflow with Biofidelic Sideways Fall FE...

Sophisticated biofidelic finite element (FE) models of sideways falls are an emerging tool for predicting hip fracture risk. We adapted an existing experimental setup for in-silico trials by creating an automated workflow to build FE models of the experiment and characterizing it with respect to the effects of limited CT scan coverage. Limited CT scan coverage was simulated by shortening the femur (25-175mm distal to the greater trochanter), misaligning the femur up to 5 degrees, and using a morphed template pelvis. We compared impact force error (E_BL) of the FE results to the existing experimental results. We then characterized the limited CT scan coverage with respect to the impact force (E_I), femur force (E_F), and fragility ratios based on the impact force (E_FRI) and the femur force (E_FRF) by comparing results to a model without any of these scan-related errors introduced. In general, the baseline simulations agreed well with the experiments (E_BL: μ = - 0.007 kN, σ = 0.409 kN). When scan coverage errors were introduced, the errors were small (E_I: μ = 0.082kN, σ = 0.232kN; E_F: μ = 0.063 kN, σ = 0.228 kN; E_FRI: μ = - 0.00, σ = 0.058; E_FRF: μ = - 0.012, σ = 0.074). The pelvis template used explained the most variance of the output measures (E_I R² = 0.876; E_F R² = 0.880; E_FRI R² = 0.901; E_FRFR² = 0.884). These results indicate that this automated methodology is suitable for typical scan coverages encountered clinically, and future work should use this workflow to explore fragility fractures in larger clinical cohorts.