Novel Approach to Predict Atrial Fibrillation (AF) Recurrence after Pulmonary Vein Isolation using Simulations of Patient-Specific MRI Models and Machine Learning

Unmet Need
Atrial fibrillation (AF) is an irregular, rapid heart rate in which the two upper chambers (atria) of the heart beat chaotically and out of sync with the lower chambers (ventricles) of the heart. Ultimately, this condition can increase one’s risk of heart disease, stroke, and other cardiovascular complications. AF is thus regarded as an important healthcare problem as it results in more than 750,000 hospitalizations and 130,000 deaths each year in the United States, generating $6 billion in healthcare costs. It is estimated that 2.7-6.1 million people in the United States are currently living with AF. One procedure used to treat AF in patients that have fibrosis in the atria is catheter ablation in which the pulmonary veins are isolated with catheter ablation lesions. However, 20-40% of patients often experience recurrent AF following the first ablation procedure and must undergo a repeat procedure, which presents additional complications and considerations. Consequently, there is an unmet need for a method that can predict the likelihood of recurrence following ablation and pulmonary vein isolation in patients and can aid in the suggestion of strategic, individualized approaches for more successful index AF ablation surgeries.
 
Technology Overview
Johns Hopkins researchers have developed a novel approach that can be used to predict AF recurrence after ablation surgery and pulmonary vein isolation. This approach constructs an individualized, patient-specific model of the patient’s atria using a contrast-enhanced MRI along with stimulations to characterize the types of arrhythmia that could possibly occur. The results from the stimulations and MRI are then fed into a machine learning algorithm to determine the set of features that is the most predictive of AF recurrence following the ablation procedure. The algorithm will have the ability to predict with relatively high accuracy, whether isolating the pulmonary vein will successfully eliminate the patient’s AF or if the physician must perform additional ablation in the fibrotic parts of the atria.
 
Stage of Development
The inventors have developed a novel method and machine learning algorithm to predict the outcomes and best methodology to be used in ablation procedures for patients with AF in order to minimize complications and prevent recurrence. At present, additional research and development is required to optimize and train the algorithm for widespread use in patients with AF.
 

Category(s):