Novel Framework to Integrate Real-Time MR-Guided EP Data with T1 Mapping-Based Computational Heart Models

  • Sebastian Ferguson
  • Maxime Sermesant
  • Samuel Oduneye
  • Sophie Giffard-Roisin
  • Michael Truong
  • Labonny Biswas
  • Nicholas Ayache
  • Graham Wright
  • Mihaela Pop
Conference paper

DOI: 10.1007/978-3-319-52718-5_2

Part of the Lecture Notes in Computer Science book series (LNCS, volume 10124)
Cite this paper as:
Ferguson S. et al. (2017) Novel Framework to Integrate Real-Time MR-Guided EP Data with T1 Mapping-Based Computational Heart Models. In: Mansi T., McLeod K., Pop M., Rhode K., Sermesant M., Young A. (eds) Statistical Atlases and Computational Models of the Heart. Imaging and Modelling Challenges. STACOM 2016. Lecture Notes in Computer Science, vol 10124. Springer, Cham

Abstract

Real-time MRI-guided electrophysiology (EP) interventions hold the potential to replace conventional X-ray guided procedures aimed to eliminate potentially lethal scar-related arrhythmia. Furthermore, although cardiac MR can provide excellent structural information (i.e., anatomy and scar), these catheter-based procedures have limited electrical information due to sparse electrical maps recorded from endocardial surfaces. In this paper, we propose a novel framework to augment such sparse electrical maps with 3D transmural electrical wave propagation obtained non-invasively using computer modelling. First, we performed real-time MR-guided EP studies using a preclinical pig model (i.e., in 1 healthy and 2 chronically infarcted animals). Specifically, the MR scans employed 2D T1-mapping (1 × 1 × 5 mm spatial resolution) based on a multi-contrast late enhancement method. For the EP studies we used an MR-compatible system (Imricor). Second, the stacks of resulting segmented images were used to build 3D heart models with various zones (i.e., healthy, scar and gray zone). Lastly, the 3D heart models were coupled with simple monodomain reaction-diffusion equations (e.g. eikonal and Aliev-Panfilov). Our simulations showed that these mathematical formalisms are advantageous due to fast computations, allowing us to predict the electrical wave propagation through dense LV meshes (e.g. >100 K elements, element size ~1.5 mm) in <3 min on a consumer computer. Overall, preliminary results demonstrated that the 3D MCLE-based models predicted close activation times and patterns compared to our measured EP maps, while also providing 3D transmural information and a precise location of the infarction. Future work will focus on calibrating directly (in near real-time) T1-based personalized heart models from electrical maps obtained during real-time MR-guided EP mapping procedures.

Keywords

Cardiac MRI Modelling Electrophysiology Histopathology 

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  • Sebastian Ferguson
    • 1
  • Maxime Sermesant
    • 2
  • Samuel Oduneye
    • 1
    • 3
  • Sophie Giffard-Roisin
    • 2
  • Michael Truong
    • 1
    • 4
  • Labonny Biswas
    • 1
  • Nicholas Ayache
    • 2
  • Graham Wright
    • 1
    • 3
  • Mihaela Pop
    • 1
    • 3
  1. 1.Sunnybrook Research InstituteTorontoCanada
  2. 2.Inria - Asclepios ProjectSophia AntipolisFrance
  3. 3.Medical BiophysicsUniversity of TorontoTorontoCanada
  4. 4.Kings College LondonLondonUK

Personalised recommendations