CAREER: A Parallel Computational Framework of Multiscale Geometric Modeling and Mesh Generation for Cardiac Biomechanics Application

该项目的总体目标是建立一种可应用于心脏生物力学应用的多尺度几何建模和网格生成的新型并行计算框架。这一能力将使人们能够准确、稳定、高效地模拟心肌细胞内钙离子介导的信号传递、兴奋收缩偶联和能量代谢等多种生物学过程，从而推动心脏生物力学的研究取得重大进展。在心肌细胞中，钙离子最为人所知的是它在收缩激活中的作用。钙离子分布的改变现在被认为是包括心律失常和肥厚在内的各种常见病理中心功能不全的主要机制。为了预测和分析钙动力学和心肌兴奋-收缩偶联是如何被调节的，需要对大型、复杂的T小管网络和相关的蛋白质复合体的真实几何形状进行建模。然而，以前的研究仅限于简化的几何图形。在这个项目中，PI主要集中在1)从蛋白质数据库中的原子分辨率数据开始对蛋白质复合体进行多尺度几何建模；2)解决拓扑模糊和曲率驱动质量改进的并行网格生成；3)复杂几何形状的心室肌细胞钙信号自适应有限元分析中的模型验证。为了处理如此大型、复杂的系统，将开发多核并行啮合工具包，并将其与仿真软件封装在一起。所提出的研究将在模型开发和仿真中达到最高的精度、效率和稳健性。它将显著提高心脏应用的预测能力，以及在分子和细胞尺度上对解剖和生理特性的理解。这种并行计算基础设施还可以用于其他复杂系统，为工程师和科学家提供构建准确计算机模型的新技术。此外，这一跨学科项目将通过新的教育工具和课程开发以及外联活动，将研究和教育结合起来。学生将与合作机构互动，以获得真实问题的第一手经验。妇女、少数群体和高中生将通过CMU的K-12项目被纳入拟议的研究和教育活动。教育活动将与CMU的Eberly卓越教学中心一起进行评估。

The overall goal of this CAREER project is to establish a novel parallel computational framework of multiscale geometric modeling and mesh generation that can be applied to cardiac biomechanics applications. This capability will enable accurate, stable, efficient simulations of many biological processes such as calcium (Ca2+) mediated signaling, excitation-contraction coupling and energy metabolism in cardiac muscle cells, and lead to great advances in cardiac biomechanics. In cardiac muscle cells, Ca2+ is best known for its role in contraction activation. Alterations in Ca2+ distributions are now recognized to be the primary mechanisms of cardiac dysfunction in a diverse range of common pathologies including cardiac arrhythmias and hypertrophy. To predict and analyze how Ca2+ dynamics and cardiac excitation-contraction coupling are regulated, modeling of realistic geometries of large, complicated t-tubule network and associated protein complexes is needed. However, previous studies have been limited to simplified geometries. In this project, the PI focuses on 1) multiscale geometric modeling for protein complexes starting from atomic resolution data in the Protein Data Bank; 2) parallel mesh generation with topology ambiguity resolved and curvature-driven quality improvement; and 3) model validation in adaptive finite element analysis of Ca2+ signaling in ventricular myocytes with complicated realistic geometry. To handle such large, complicated systems, multicore parallel meshing toolkits will be developed and encapsulated with the simulation software. The proposed research will attain the highest degree of accuracy, efficiency and robustness in model development and simulation. It will significantly advance predictive capability in cardiac applications, and the understanding of anatomical and physiological properties at molecular and cellular scales. This parallel computational infrastructure can also be used for other complicated systems, providing engineers and scientists with novel technologies to construct accurate computer models. Furthermore, this interdisciplinary project will integrate research and education via novel educational tool and curriculum development as well as outreach activities. Students will interact with collaborative institutions to gain firsthand experience of real issues. Women, minority groups and high school students will be included in the proposed research and education activities through CMU's K-12 Programs. Education activities will be assessed in conjunction with CMU's Eberly Center for Teaching Excellence.