Project Summary Atrial Fibrillation is the most common sustained cardiac arrhythmia, and it is a major risk factor for ischemic stroke. This is due to thrombus formation in the left atrium which may be mitigated by anticoagulation therapy but carrying its own bleeding risk. Current clinical therapy selection strategies balancing bleeding risk of anticoagulation therapy to stroke risk lack for stroke patient The be prediction predictive power. Blood stasis is a key component of Virchow's triad thrombus formation, and flow imaging methods have shown lowered left atrial velocities are associated with risk. However, these methods either lack omprehensive 3D measurement or must reject data when the is i n arrhythmia, where both of these may be necessary to develop accurate biomarkers to predict stroke. goal of this project is to further develop and optimize an arrhythmia resolved flow imaging MRI method to able to capture the time varying 3D atrial hemodynamics during arrhythmia with the potential to develop stroke biomarkers to guide medical therapy selection. c The first aim of this proposal is development and optimization of the MRI method: arrhythmia resolved 5D flow MRI, a multi-dimensional data set of dimension: 3D spatial + cardiac time + heartbeat duration. For several heartbeat durations measured during arrhythmia (e.g. short heartbeats, long heartbeats) we will construct a 3D time varying magnitude and velocity data set for one cardiac cycle for that heartbeat duration. A pulsatile flow phantom, capable of simulating a known arrhythmia, will allow the optimization of acquisition time and image reconstruction methodology and parameters. I will develop two different reconstruction algorithms, which can be reconstructed using the same raw data. First, images will be reconstructed to measure maximum differences across the varying heart rhythm in order to analyze the heart rate variable hemodynamics within a patient. Second, images will be reconstructed based on data acquired from fixed, standardized heartbeat lengths, allowing the comparison between subjects, controlling for the effect of heartbeat duration itself. This will be tested against the reference standard, in this setting: 2D real time phase contrast. Clinical collaborators will help with algorithm design of the optimized imaging reconstruction to ensure success of in vitro to in vivo translation of the technique from the first aim to subsequent aims. The second aim will validate this technique in vivo with a cohort of 10 healthy controls and 10 persistent atrial fibrillation patients. This will also establish heart rate variable and controlled hemodynamic differences between these groups. In the third aim, we will test the abilities of the method to discriminate between stroke history and no stroke history atrial fibrillation patients. 12 persistent atrial fibrillation patients with prior cardioembolic stroke history and 12 persistent atrial fibrillation patients without a stroke hist...