PROJECT SUMMARY Sudden cardiac death due to ventricular arrhythmias (VAs) is a leading cause of mortality world-wide. Despite advancements in anti-arrhythmia therapies, VA rates remain high in part due to an incomplete understanding of the underlying disease-induced arrhythmigenic substrate. In patients with prior myocardial infarction, traditional dogma has maintained that heterogeneous scarring in the ventricles forms the arrhythmia substrate. While clinical studies have utilized the visualization of ventricular scar in localizing VA ablation targets, these efforts have failed to significantly improve VA recurrence rates, suggesting that scar characterization alone may be insufficient for identifying and eliminating VA. Infiltrating adipose tissue (inFAT) is a newly recognized aspect of post-infarct remodeling. However, because inFAT is intermingled with scar, the specific role of lnFAT In VA propensity In patients with ischemic cardiomyopathy has never been explored. The overall obJectlve of this appllcatlon Is to use a novel combination of mechanistic personallzed computatlonal modellng ("dlgltal twin" of the heart), Imaging, electroanatomlcal mapping, ex-vivo human heart experiments, and artificial intelligence (Al) to comprehensively characterize the role of inFAT vs. scar in post-infarct VAs, and to develop a new digital-twin approach for guidance of VA ablation in patients with ischemic cardiomyopathy. Leveraging our advancements in the acquisition of high-quality ventricular images of scar and inFAT distribution, our expertise in personalized computational modeling and Al, and our clinical and experimental expertise, we propose to develop personalized heart digital twins of ICM patinets that incorporate scar and inFAT distributions and are parameterized with experimental data. Using the digital twins and intra-procedure data, we will explore the mechanistic role of inFAT in arrhythmogenic propensity and in the components of the VA circuit. We will also utilize the digital twinning technology to develop a comprehensive VA ablation guidance strategy that accounts for the roles of inFAT and scar in VA circuits. The project will culminate in a clinical translation feasibility study to demonstrate that the novel digital twins offer accurate prediction of VA ablation sites, and can be used for pre-procedure guidance and optimal targeting, eliminating extensive electroanatomical mapping. Successful execution of the proposed studies will provide new mechanistic understanding of the role of inFAT in promoting and sustaining VAs, and will lead to significant improvements in the clinical procedure of VA ablation. Completion of this project will also be a major leap forward in the integration of imaging, computational modeling, intracardiac mapping, and Al in the treatment of heart rhythm disorders.