Efficient Formation of Patient-Specific Finite-Element Models of the Heart
Finite Element (FE) simulation has been extensively used in cardiac modeling due to its ability to replicate normal cardiac physiology and adapt to simulate various cardiac pathologies. As such, different FE cardiac models have been developed, but representing only a limited number of anatomical variations. A major bottleneck in the creation of new geometries, as with most anatomically based models, is image segmentation. In this work, we demonstrate a proof-of-concept workflow leveraging the Cardiac CT Auto Segmentation tool in the Synopsys' Simpleware™ software (Synopsys Inc., Mountain View, CA) [8]. This workflow aims to quickly and automatically segment patient anatomy from 4D cardiac CT image data and create patient-specific FE models. In our study, 4D CT data with 10 time steps through the entire cardiac cycle were imported into Synopsys' Simpleware, and the Cardiac CT Auto Segmentation tool segmented each timeframe in a fully automated process. The segmented patient 4D CT data can thus generate new FE cardiac geometries. Cardiac meshes produced from Synopsys' Simpleware's segmentations can be converted into a LS-DYNA model for patient-specific modeling and FE analysis. As demonstrated, our approach significantly reduces the manual effort required for image segmentation, facilitating the creation of clinically relevant and personalized cardiac models.
