Semi-automatic tool to identify heterogeneity zones in lge-cmr and incorporate the result into a 3d model of the left ventricle

Maria Narciso, António Ferreira, Pedro Vieira

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

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.

Original languageEnglish
Title of host publicationImage Analysis and Recognition - 17th International Conference, ICIAR 2020, Proceedings
EditorsAurélio Campilho, Fakhri Karray, Zhou Wang
Place of PublicationCham
PublisherSpringer
Pages238-246
Number of pages9
ISBN (Electronic)978-3-030-50516-5
ISBN (Print)978-3-030-50515-8
DOIs
Publication statusPublished - 2020
Event17th International Conference on Image Analysis and Recognition, ICIAR 2020 - Póvoa de Varzim, Portugal
Duration: 24 Jun 202026 Jun 2020

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PublisherSpringer
Volume12132 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference17th International Conference on Image Analysis and Recognition, ICIAR 2020
CountryPortugal
CityPóvoa de Varzim
Period24/06/2026/06/20

Keywords

  • Arrhythmia
  • Gray zone
  • Heart computational model
  • Ischaemia

Fingerprint

Dive into the research topics of 'Semi-automatic tool to identify heterogeneity zones in lge-cmr and incorporate the result into a 3d model of the left ventricle'. Together they form a unique fingerprint.

Cite this