Image-Based Computational Modelling of Fetal and Adult Cardiac Biomechanics
Dr Choon Hwai Yap (Imperial College London)
Thursday 21st October 14:00-15:00 Room 110/ZOOM (ID: 994 9050 0701)
Congenital heart malformations lead to many pregnancy terminations and occurs to about 1% of births, and is the leading cause of birth defect related deaths. There is evidence that biomechanical abnormalities can lead to malformations. Some malformations are caused by cardiac abnormalities during mid-gestation that disrupts the biomechanical environment. In such cases, catheter-based fetal heart interventions have shown promise in resolving the abnormalities and preventing progression to malformation at birth. Our lab is interested in developing an understanding of the biomechanics of the fetal heart, to understand and predict fetal progression of congenital heart diseases, and to predict outcomes of fetal heart interventions, so as to more accurately assess suitability of the procedure for patients.
Towards this goal, we first developed enabling tools, including a cardiac motion estimation algorithm to track 3D cardiac motions from noisy echocardiography images and extract 3D strains, and an algorithm to compound 3D echo images across the entire cardiac cycle for image enhancements. With image-based flow simulations, we characterized the normal fluid dynamics in the fetal heart, and elucidated the abnormal flow during fetal aortic stenosis and evolving Hypoplastic Left Heart Syndrome. With image-based finite element simulations, we characterized the abnormal biomechanics expected of fetal aortic stenosis, and developed a methodology to simulate the post-interventional conditions based on pre-interventional data, paving the way for a tool to predict fetal heart intervention outcomes.
We are also in applying our tools to understand the biomechanics of adult heart failure, as heart failure remain the top reason for global mortality, in hope of inspiring novel diagnostic and therapeutic tools. For example, we performed image-based finite element modelling of a porcine HFpEF model, produced by progressive aortic constriction, and tracked disease progression over 6 weeks. We found that active tension and myofiber stresses elevated quickly at the onset, followed by increased tissue stiffness and myocardium thickness and decreased longitudinal strains two weeks later. Tissue stiffness and myocardium thickness were significantly correlated with myofiber stress, suggesting a possible causative effect. Closer analysis of the left ventricular (LV) geometry versus biomechanical function showed that LV geometry consequent to heart failure remodeling has significant effects on cardiac function, and can skew well-used clinical parameters such as the ejection fraction. However, a corrected ejection fraction parameter calculated with the mid-wall layer instead of endocardial boundary removes this skewedness and enable distinguishing between HFpEF and HFrEF hearts.