Figure 1. Overview of the approach and geometrical model reconstruction.
(A) Flowchart summarizing the virtual-heart ventricular arrhythmia (VA) risk stratification approach for hypertrophic cardiomyopathy (HCM) patients. A combination of late gadolinium enhancement (LGE)-cardiac magnetic resonance (CMR) and postcontrast T1 mapping is used to construct personalized left ventricular (LV) geometrical models with fibrotic remodeling. Incorporating HCM-specific electrophysiological properties (action potential kinetics, conduction velocity) completes the generation of each personalized LGE-T1 virtual heart, which is then used to assess VA propensity in the substrate via rapid pacing. right ventricle (RV) is shown in transparent gray. Dense fibrosis (scar) is considered nonconductive. (B) Fusing LGE-CMR and postcontrast T1 map information to generate the personalized, geometrical virtual-heart model. Top left: LV segmentation with intermediate and high signal intensity thresholds of 3 (yellow) and 5 standard deviation (SD) (purple), respectively, on short-axis LGE-CMR. Bottom left: mid-ventricular postcontrast T1 map segmentation with relaxation time thresholds of <350 (blue) and 350–450 ms (gray). Right: the thresholds of the LGE-CMR signal intensity were adjusted to new, personalized thresholds, TDiffuse and TDense, based on the T1 map (see text for detail). The new personalized signal intensity thresholds in the matching LGE-CMR slice were then applied to all LGE-CMR short-axis slices.