New AI Research Helps Predict Likelihood of AFib

Researchers with the Emory University Empathetic AI for Health Institute authored the article, “Artificial Intelligence–Based Feature Analysis of Pulmonary Vein Morphology on Computed Tomography Scans and Risk of Atrial Fibrillation Recurrence After Catheter Ablation: A Multi-Site Study,” which was recently published in the American Heart Association Journal. 

The study evaluated whether an artificial intelligence model using morphology could predict if atrial fibrillation (AF+) would occur after a common heart procedure, catheter ablation.  

AI models were trained to analyze CT images, allowing for the segmentation and assessment of primary and secondary pulmonary vein (PV) branches. Data from more than 800 patients from three institutions were used to build and test classifiers based on 60 quantitative features. 

The results revealed that primary PV branches were more closely linked to AF recurrence than secondary PV branches. The AI models, especially the primary PV model (Mp), accurately predicted AF+ cases, with fractal dimension—a measure of surface complexity—playing a key role.

The study concluded that the morphology of the primary PV branches is a critical factor in predicting AF recurrence after ablation.

To view the research in its entirety, click the link below below:

https://www.ahajournals.org/doi/abs/10.1161/CIRCEP.123.012679