Identifying personalized parameters for left ventricle model of the heart
- PDF / 14,744,625 Bytes
- 13 Pages / 439.37 x 666.142 pts Page_size
- 114 Downloads / 197 Views
Identifying personalized parameters for left ventricle model of the heart Raheem Gula , Muniba Javaid, Aamir Shahzad COMSATS University Islamabad, Abbottabad Campus, Pakistan Received: 26 April 2020 / Accepted: 13 September 2020 © Società Italiana di Fisica and Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract Patient-specific computational models of the human heart are important in understanding the heart hemodynamics which eventually help in improving clinical treatments. In order to develop a personalized model of the human heart, it is essential to identify personalized heart parameters which further can be measured directly or indirectly from the patient data or measurements. Within this work, we considered an existing low-dimensional (0D) model of the left ventricle (LV) and applied global sensitivity analysis (Sobol’s method) to identify the personalized heart parameters. Based on the results of sensitivity analysis, we observed that the identification of personalized heart parameters for all output quantities of interest is different. A complete summary of personalized heart parameters for each output quantities of interest is given at the end of this paper. The study is useful for experimentalists to develop patient-specific/personalized model(s) of LV by measuring/estimating personalized (key) heart parameters in complement with the patient data.
1 Introduction Personalized or patient-specific models of human cardiovascular system (CVS) are important tools to understand the hemodynamics in healthy and diseased individuals. Recent advancements in mathematical modeling and medical image processing have had an important impact in the development of diagnosis and treatment of cardiovascular diseases (CVD). In patientspecific modeling, researchers try to find a good agreement between model simulations and patient(s) data, which eventually improve the quality of medical diagnosis and treatments. From the last few decades, different flavors of computational models of the CVS, i.e., 3D, 2D, 1D, 0D and multi-dimensional are developed. Out of those, 0D models of CVS are simple, easy to implement, computationally efficient and potentially beneficial for real-time simulations if compared with high or multi-dimensional models [11–18]. As 0D models of CVS are computationally less expensive, they are suitable for uncertainty quantification, sensitivity analysis and parameter estimation. For example, Duanmu, Z. [2] and Kim, H. J. et al. [1] developed patient-specific model of whole coronary circulation in complement with the CT scans data taken from patient-specific geometries. Tache, I. A. et al. [3] coupled 0D model of the heart with coronary circulation to provide the input boundary condition at the inlet of the main aorta. Further, Tylor, C.A. et al. [5] discussed some progress in the
a e-mail: [email protected] (corresponding author)
0123456789().: V,-vol
123
763
Page 2 of 13
Eur. Phys. J. Plus
(2020) 135:763
field of numerical methods and 3D imaging techniques to develop patient-s
Data Loading...
