Uncertainty Quantification of Cardiac Properties | 85 This protocol ran for at least 500 iterations. Additional iterations were performed in the case that the effective sample size Neff >10 ⋅ nθ >10 Neff >10 ⋅ nθ was not fulfilled. The Kish effective sample size Neff was used15,which is defined as Neff = [∑θ∈Θ w(θ)] 2 ∑θ∈Θ(w(θ)2) . (6) Problem description Clinical data Patient-specific simulations were based on echocardiographic data from AC mutation carriers in various disease stages. Besides clinically measured LV and RV regional deformation imaging data, the LV end-diastolic volume (EDV), LV ejection fraction (EF) and right ventricular basal diameter (RVD) were used as model input. We used echocardiographic data of nine pathogenic AC mutation carriers which were evaluated in the University Medical Center Utrecht, the Netherlands. As previously described4, deformation analyses of these echocardiograms were performed twice by two observers to determine clinical inter- and intra-observer variability. Lastly, longitudinal datasets with >2 echocardiograms per patient at different time points were used to explore applicability of the model for follow-up of tissue properties over time. These longitudinal datasets were acquired from AC mutations carriers which were evaluated in the Oslo University Hospital, Norway. All echocardiographic data were obtained on a Vivid 7, Vivid 9 or Vivid E95 ultrasound machine (GE Vingmed, Horten, Norway). The echocardiographic protocol was described previously.16 In this study, we focused on the right ventricular free wall (RVfw). This is typically the most affected area in AC mutation carriers17, which is expressed in typical deformation abnormalities (delayed onset of shortening, decreased peak systolic strain, post-systolic shortening, and increased RV mechanical dispersion).16 Therefore, deformation patterns of three RVfw segments (apical, mid-ventricular and basal) were used as input for our modelling framework (Figure 1).4 Additionally, LV free wall (LVfw) and interventricular septal (IVS) deformation patterns were included to ensure realistic mechanical boundary conditions for the RVfw in terms of ventricular interaction. These patterns were obtained by averaging the 12 LVfw and 6 IVS segmental deformation curves, respectively, using the standardized 18-segment model.18 Computational model of heart and circulation Clinical measures were simulated using the CircAdapt model. This model is a fast biomechanical lumped parameter model of the heart and circulation. Via the one fibre model19, wall stress is related to cavity pressure. The TriSeg module allows inter-ventricular interaction over the IVS.20 Phenomenological material laws prescribe the stress-strain relation in the spherical walls. The MultiPatch module allows for regional heterogeneity of tissue properties within a single wall21 and is used to describe the heterogeneity in the RVfw. Three segments were created in the RVfw to model the mechanics in the three different RVfw segments (apical, mid-ventricular, and basal). 5
RkJQdWJsaXNoZXIy MTk4NDMw