A Multi-Scale Computational Model for the Rat Ventricle: Construction, Parallelization, and Applications

Xiangpeng Bi,Shugang Zhang,Huasen Jiang,Zhiqiang Wei
DOI: https://doi.org/10.1016/j.cmpb.2021.106289
2021-01-01
Abstract:Background: Cardiovascular diseases are the top killer of human beings. The ventricular arrhythmia, as a type of malignant cardiac arrhythmias, typically leads to death if not treated within minutes. The multi-scale virtual heart provides an idealized tool for exploring the underlying mechanisms, by means of incor-porating abundant experimental data at the level of ion channels and analyzing the subsequent patholog-ical changes at organ levels. However, there are few studies on building a virtual heart model for rats-a species most widely used in experiments. Objective: To build a multi-scale computational model for rats, with detailed methodology for the model construction, computational optimization, and its applications. Methods: First, approaches for building multi-scale models ranging from cellular to 3-D organ levels are introduced, with detailed descriptions of handling the ventricular myocardium heterogeneity, geometry processing, and boundary conditions, etc. Next, for dealing with the expensive computational costs of 3-D models, optimization approaches including an optimized representation and a GPU-based parallelization method are introduced. Finally, methods for reproducing of some key phenomenon (e.g., electrocardio-graph, spiral/scroll waves) are demonstrated. Results: Three types of heterogeneity, including the transmural heterogeneity, the interventricular het-erogeneity, and the base-apex heterogeneity are incorporated into the model. The normal and reentrant excitation waves, as well as the corresponding pseudo-ECGs are reproduced by the constructed ventricle model. In addition, the temporal and spatial vulnerability to reentry arrhythmias are quantified based on the evaluation experiments of vulnerable window and the critical length. Conclusions: The constructed multi-scale rat ventricle model is able to reproduce both the physiological and the pathological phenomenon in different scales. Evaluation experiments suggest that the apex is the most susceptible area to arrhythmias. The model can be a promising tool for the investigation of arrhythmogenesis and the screening of anti-arrhythmic drugs. (c) 2021 Elsevier B.V. All rights reserved.
What problem does this paper attempt to address?