Critical to improving our understanding and treatment of cardiac pathologies is the understanding of the basic mechanisms that underpin cardsic function. Multi-scale biophysical models quantatively link sub-cellular protein function to whole organ physiology. Computational models assist in combining and interpreting disparate experimental data sets recorded across multiple modalities and research groups into a common representation of the heart. Combining detailed measurements within a physically constrained framework allows us to investigate the effects of genetic manipulation, aging and disease progression from the protein through to pump function.
Atrial fibrillation (AF) is a supraventricular tachyarrhythmia characterised by uncoordinated atrial activation with consequent deterioration of mechanical function. AF is the most common arrhythmia, affecting almost 1 million people in the UK, and increases the risk of other cardiovascular diseases, stroke and death. Catheter ablation is routinely used to treat AF in drug-refractory patients, but has moderate efficacy (only 50-75% long term maintenance of sinus rhythm). We aim to use personalised patient specific models to improve our understanding of the mechanisms behind AF to improve patient selection, disease management and procedure planning.
Heart failure is a progressive and prevalent disease. Cardiac resynchronisation therapy (CRT), where pacing electrodes are attached to the heart to improve electrical activation, remains one of the few and effective treatments for CRT. However, 30-50% patientís hearts do not reverse remodel or revert back to a healthy physiology. We have developed patient specific models of patients hearts who receive CRT to analyse the effects of pacing on cardiac mechanics, to evaluate device design and determine the role of cellular function on patient response.
In the US up to 31% of drug withdrawals are attributed to cardiac toxicity. Many drugs pass through preclinical trials without indication of potential cardio-toxic risk. To improve identification of cardiotoxicity we propose to develop computational models that quantitatively analyse the impact of drugs on cellular and whole organ function. Multi-scale models that integrate protein scale drug effects within a cell to whole organ electrical and mechanical function provide a novel and comprehensive method for evaluating drug cardiotoxicity.
The shift of cardiac models from a research tool to a potential clinical product for informing patient care will bring cardiac models into the remit of clinical regulators. With this transition comes the requirement for improved coding standards, and an increased interest in Verification, Validation, and Uncertainty Quantification (VVUQ). We have led efforts to create standard benchmarks for verification of cardiac electrophysiology and cardiac mechanics. A description of the cardiac mechanics benchmark problems, published solutions, and scripts for mesh generation can be found here.