Semi-automatic Tool to Identify Heterogeneity Zones in LGE-CMR and Incorporate the Result into a 3D Model of the Left Ventricle
- 140 Downloads
Fatal scar-related arrhythmias are caused by an abnormal electrical wave propagation around non conductive scarred tissue and through viable channels of reduced conductivity. Late gadolinium enhancement (LGE) cardiovascular magnetic resonance (CMR) is the gold-standard procedure used to differentiate the scarred tissue from the healthy, highlighting the dead cells. The border regions responsible for creating the feeble channels are visible as gray zones. Identifying and monitoring (as they may evolve) these areas may predict the risk of arrhythmias that may lead to cardiac arrest. The main goal of this project is the development of a system able to aid the user in the extraction of geometrical and physiological information from LGE images and the replication of myocardial heterogeneities onto a three-dimensional (3D) structure, built by the methods described by our team in another publication, able to undergo electro-physiologic simulations. The system components were developed in MATLAB R2019b the first is a semi-automatic tool, to identify and segment the myocardial scars and gray zones in every two-dimensional (2D) slice of a LGE CMR dataset. The second component takes these results and assembles different sections while setting different conductivity values for each. At this point, the resulting parts are incorporated into the functional 3D model of the left ventricle, and therefore the chosen values and regions can be validated and redefined until a satisfactory result is obtained. As preliminary results we present the first steps of building one functional Left ventricle (LV) model with scarred zones.
KeywordsGray zone Ischaemia Arrhythmia Heart computational model
- 7.Narciso., M., Sousa., A.I., Crivellaro., F., de Almeida., R.V., Ferreira., A., Vieira., P.: Left ventricle computational model based on patients three-dimensional MRI. In: Proceedings of the 13th International Joint Conference on Biomedical Engineering Systems and Technologies, vol. 2 BIOIMAGING: BIOIMAGING, pp. 156–163. INSTICC, SciTePress (2020). https://doi.org/10.5220/0008961601560163