Characterising Atrial Anisotropy and Fibrosis For Patient-Specific Models of Atrial Fibrillation Ablation

Lead Supervisor
Professor Steven Niederer
School of Biomedical Engineering & Imaging Sciences, King’s College London
steven.niederer@kcl.ac.uk

Co-supervisor
Dr Martin Bishop
Reader in Computational Cardiac Electrophysiology
School of Biomedical Engineering & Imaging Sciences, King’s College London

Industrial Partners / Collaborators
Professor Mark O’Neill, Dr John Whitaker, Professor Mike Shattock and Dr Caroline Roney¬†

Project Details

Atrial fibrillation (AF) is the most common cardiac arrhythmia, affecting over 1.1 million people in the UK alone, and is associated with increased risk of cardiovascular diseases, stroke and death. Radio-frequency catheter ablation therapy, which creates lesions in cardiac tissue, is a routine treatment for drug refractory AF. Early in AF, isolating the pulmonary veins (PVI) using catheter ablation eliminates electrical triggers and is a successful treatment approach. With increasing AF duration, the atrial tissue undergoes electrical and structural fibrotic remodelling, wavefront propagation becomes chaotic, and PVI catheter ablation treatment has a lower success rate. These persistent AF patients are a heterogeneous population: some patients require multiple procedures, with more extensive ablation strategies; while for others, PVI is sufficient, as indicated by the recent STAR AF II clinical trial. Identifying persistent AF patients where PVI will be a sufficient treatment remains a clinical challenge, which if solved could lead to improved safety, better patient selection, as well as decreased time and cost for procedures. 

Previous clinical and modelling studies suggest that the amount and location of atrial fibrosis affects arrhythmia properties and ablation outcome. In particular, many of the changes that occur during AF that modify atrial conduction are associated with atrial fibrosis, including the deposition of collagen and interstitial fibrosis, as well as changes in atrial fibre direction, including fibre disarray. These components of fibrotic remodelling affect the heterogeneity and anisotropy of atrial conduction. The interpretation of atrial conduction properties and their relationship with scar tissue requires the use of atrial anisotropy and fibre direction information. However, currently atrial fibre direction cannot be recorded in vivo and atrial electrical anisotropy is unknown. Characterising the atrial substrate scar and conduction properties may inform ablation approaches through improved understanding of the arrhythmia substrate or through patient-specific modelling.  The first aim of this project is to characterise atrial activation during pacing to construct an anisotropy atlas for healthy patients, together with a measure of variability. The second aim is to investigate the relationship between fibrotic remodelling and atrial conduction properties. Finally, this information will be used to construct patient-specific models of AF patients to predict PVI ablation outcome. This project will provide training in computational modelling, signal and image processing techniques, and machine learning algorithms, applied to the field of cardiac electrophysiology. Specifically, the project involves analysing imaging and electrical data from persistent AF patients, in close collaboration with the clinical teams at GSTT. These data will be used to tune biophysical models to investigate the effects of electrical and structural properties on AF. 

Datasets

Work will rely electrical data collected as part of an ongoing clinical atrial mapping study and using the GSTT atrial MRI data set

Keywords

Electrophysiology, Atria, Simulation, Machine Learning