A Digital Twin Approach for Stroke Risk Assessment in Atrial Fibrillation Patients

Patients with atrial fibrillation (AF) report a fivefold increased risk of cerebrovascular events, being responsible of 15–18 % of all strokes. In clinical practice stroke prevention is based on the
$\text{CHA}_{2}\text{DS}_{2}$
- VASc score mainly used to identify low-risk patients and a quantitative index for stroke risk stratification is missing. In this study we propose a digital twin (DT) model of the left atrium (LA) and the use of computational fluid dynamic simulations (CFD) to improve stroke risk assessment on a patient-specific basis. Simulations were performed on patient-specific dynamic LA models in sinus rhythm (SR) in two subjects (one AF patient and one normal control). Blood velocity, regions prone to thrombogenesis through the computation of endothelial damage, and blood stasis were assessed in the two subjects, showing different profiles (higher velocities, lower endothelial damage and decreased blood stasis in the control subject) in both the LA chamber and the left atrial appendage. Following testing on a larger population and a classification analysis, the proposed approach may be used to improve stroke risk assessment.