Predicting atrial fibrillation recurrence after pulmonary vein isolation using simulations of patient-specific magnetic resonance imaging models and machine learning

A device, comprising:

• one or more memories;
• and
• one or more processors, communicatively coupled to the one or more memories, configured to:
• receive clinical late gadolinium enhanced cardiac magnetic resonance imaging (LGE-MRI) images of a patient;
• perform segmentation of epicardial and endocardial surfaces on the LGE-MRI images to create a patient-specific three-dimensional (3D) model for the patient;
• identify normal tissue regions and atrial fibrosis regions in the 3D model based on intensity;
• divide the 3D model into the normal tissue regions and the atrial fibrosis regions;
• assign first cell and tissue properties to the normal tissue regions according to a model of human cardiac myocyte and based on transverse and longitudinal conduction velocities;
• assign second cell and tissue properties to the atrial fibrosis regions according to a model of diffuse fibrosis with complete fibrotic remodeling;
• perform simulations on the normal tissue regions and the atrial fibrosis regions, based on the first cell and tissue properties and the second cell and tissue properties, to generate simulation results;
• extract first features from the simulation results based on the 3D model and based on a simulation protocol;
• extract second features from the LGE-MRI images;
• process the first features and the second features, with a machine learning model, to select a feature, from at least one of the first features or the second features, that is predictive of atrial fibrillation recurrence;
• and
• utilize the feature to augment the simulation protocol for a future patient.

A non-transitory computer-readable medium storing instructions, the instructions comprising:

• one or more instructions that, when executed by one or more processors, cause the one or more processors to:
• receive clinical late gadolinium enhanced cardiac magnetic resonance imaging (LGE-MRI) images of a patient;
• perform segmentation of epicardial and endocardial surfaces on the LGE-MRI images to create a patient-specific three-dimensional (3D) model for the patient;
• identify normal tissue regions and atrial fibrosis regions in the 3D model based on intensity;
• divide the 3D model into the normal tissue regions and the atrial fibrosis regions;
• assign first cell and tissue properties to the normal tissue regions according to a model of human cardiac myocyte and based on transverse and longitudinal conduction velocities;
• assign second cell and tissue properties to the atrial fibrosis regions according to a model of diffuse fibrosis with complete fibrotic remodeling;
• perform simulations on the normal tissue regions and the atrial fibrosis regions, based on the first cell and tissue properties and the second cell and tissue properties, to generate simulation results;
• extract first features from the simulation results based on the 3D model and based on a simulation protocol;
• extract second features from the LGE-MRI images;
• process the first features and the second features, with a machine learning model, to select a feature, from at least one of the first features or the second features, that is predictive of atrial fibrillation recurrence;
• and
• utilize the feature to augment the simulation protocol for a future patient.