Skip to main content

Electrostructural Tomography – Towards Multiparametric Imaging of Cardiac Electrical Disorders

Periodic Reporting for period 2 - ECSTATIC (Electrostructural Tomography – Towards Multiparametric Imaging of Cardiac Electrical Disorders)

Reporting period: 2018-08-01 to 2020-01-31

Cardiac electrical diseases are directly responsible for sudden cardiac death, heart failure and stroke. They result from a complex interplay between myocardial electrical activation and structural heterogeneity. Current diagnostic strategy based on separate electrocardiographic and imaging assessment is unable to understand both these aspects. Improvements in personalised diagnostics are urgently needed as existing curative or preventive therapies (catheter ablation, multisite pacing, and implantable defibrillators) cannot be offered until patients are correctly recognised.
The aim of the ECSTATIC project is to achieve a major advance in the way cardiac electrical diseases are characterised and thus diagnosed and treated, through the development of a novel non-invasive modality (Electrostructural Tomography), combining magnetic resonance imaging (MRI) and non-invasive cardiac mapping (NIM) technologies.
This project will dramatically impact the tailored management of cardiac electrical disorders, with applications for diagnosis, risk stratification/patient selection and guidance of pacing and catheter ablation therapies. It will bridge two medical fields (cardiac electrophysiology and imaging), thereby creating a new research area and a novel semiology with the potential to modify the existing classification of cardiac electrical diseases.
Novel image-based markers of arrhythmogenicity
The main objective of ECSTATIC is to bridge cardiac imaging and cardiac electrophysiology through the development of a non-invasive approach combining information on cardiac structure and electrical activation. The first half of the project has been focused on the development of imaging and image-processing methods enabling the identification of the arythmogenic substrate responsible for cardiac arrhythmias. These new imaging methods have strong translational potential as these represent promising non-invasive markers for therapy targeting or risk stratification of sudden cardiac death.
From MR images, a novel method based on the use of free-breathing methods was found to dramatically improve the sensitivity of MRI in detecting subtle structural abnormalities in patients with ventricular arrhythmias of unknown origin. Combined with mapping methods, it allows to identify focal scars and diffuse fibrosis as independent markers of arrhythmogenicity. On the atrium, the method was employed to clarify the relationship between atrial scars and electrical reconnection of the pulmonary veins after catheter ablation. Another significant achievement has been the development of image-based models to perform patient specific simulations of cardiac electrical activation.

Novel approaches to image-based modeling and integration of non-invasive electrical measurements
Novel approaches to model personalization from heterogeneous clinical data have been developed and the ability of MRI-based models to predict patient-specific mechanisms responsible for atrial fibrillation sustenance was investigated. Last a fast and automated approach for model personalisation based on CT scan images was found able to predict electrical activation patterns within myocardial infarction. These approaches for image-based modelling can be used to perform patient-specific simulation, which can be viewed as a competitive strategy potentially alleviating the need for non-invasive body surface potential mapping.
Learning approaches based on a large simulated database were applied to the reconstruction of electrocardiography ECGI, and shown valuable for the prediction of activation patterns in dyssynchronuous heart failure. In parallel to the deep learning formulation, a novel formulation of the inverse problem aiming at identifying a source within a bidomain 3D model of the heart personalized from patient-specific images was developed. These novel approaches bridging non-invasive images and signals are at the core of the electro-structural method introduced by ECSTATIC.
The identified CT and MRI imaging markers of arrhythmogenicity can be viewed as entirely novel and should result in highly unconventional methods in the clinical management of patients with arrhythmias. Indeed, these introduce a non-invasive identification of the myocardium at risk based on 3D structural assessment, while conventional approaches rely on 2D electrical assessment based on invasive catheter measurements.
This opens new avenues for curative therapy in cardiac arrhythmias.
Image-processing approaches that provide comprehensive data on ablation targets pre-operatively could drastically improve the management and outcome of ablation procedures in patients. In ventricular arrhythmias, conventional catheter ablation approaches, largely devoted to the identification of ablation targets through catheter mapping are lengthy. It requires specific and expensive catheters, and is poorly standardized. It is reserved to expert centers and dependent upon the operator’s experience in the management of life-threatening arrhythmias and in the interpretation of complex electrical signals. Therefore, the current ablation method is often inaccurate and incomplete, leading to procedural failure characterized by arrythmia recurrence rates between 40-50% and the need for additional invasive interventions.
The imaging markers identified within ECSTATIC will allow non-invasive, 3D image-guided ablation targeting, dramatically improving the precision and efficacy of catheter ablation.

Towards electro-structural non-invasive assessment in cardiac arrhythmias
Important developments were made to make image-based patient-specific simulations of cardiac electrical activation compatible with clinical practice, both in term of robustness to data heterogeneity, and in term of computing time. With respect to non-invasive body surface potential mapping, there are currently no means to take advantage of imaging information on cardiac structure to improve the reconstruction of electrical maps from body surface measurements. Within ECSTATIC, novel formulations of the inverse problem of electrocardiography (ECGI) have been proposed, based on learning or novel source formulation. These are expected to dramatically improve the accuracy of the reconstructed maps, while at the same time enabling the interpretation of images and signals within a common framework. This may significantly improve the diagnostic/prognostic performance of non-imaging methods in patients with cardiac arrhythmias, while at the same time introducing a novel semiology at the interface between cardiac imaging and electrophysiology.
Fast personalized simulation of cardiac electrical activation from computed tomography images