MRI to MPM: Developing a Patient-specific Material Point Method Model of the Human Heart

Gavin, N.; Coombs, W.M.; Brigham, J.C.; Augarde, C.E.

Abstract

The heart is an important organ in the human body and has been widely studied using Finite Element Analysis techniques which often require extremely fine meshes in order to obtain accurate results. Also, large deformations are typically seen in heart tissue through the cardiac cycle meaning that remeshing is often required to mitigate this in FEA models. This paper covers the creation of the geometry for a new patient-specific, biomechanical model of the heart using the Material Point Method, focusing mainly on the left ventricle. The Material Point Method discretises the myocardium using material points which move through an unchanging background mesh as the heart wall deforms. In order to create the patientspecific model, a method has been developed to recreate the left ventricular geometry from segmented patient MRI scans using B-spline curves and surfaces and a least squares fitting procedure. A fitting strength parameter was introduced into the least squares fitting process to control the number of points used to create the B-spline curve and thus control the smoothness of the B-spline curve when fitting to a ring of data points from the MRI segmentation. It was seen that as the fitting strength is increased, the fitting error is reduced, with the minimum error seen when the fitting strength is at its maximum, meaning that the maximum number of points are used in the curve fitting.

Citation

Gavin, N., Coombs, W., Brigham, J., & Augarde, C. (2022, April). MRI to MPM: Developing a Patient-specific Material Point Method Model of the Human Heart. Paper presented at UKACM 2022, Nottingham, UK

• Presentation Conference Type: Conference Paper (unpublished)
• Conference Name: UKACM 2022
• Start Date: Apr 20, 2022
• End Date: Apr 22, 2022
• Deposit Date: Apr 21, 2022
• Public URL: https://durham-repository.worktribe.com/output/1136676
• Publisher URL: https://www.ukacm2022.ukacm.org/