Can AI Predict AF Recurrence After Cardioversion? | JACC Baran
カートのアイテムが多すぎます
カートに追加できませんでした。
ウィッシュリストに追加できませんでした。
ほしい物リストの削除に失敗しました。
ポッドキャストのフォローに失敗しました
ポッドキャストのフォロー解除に失敗しました
Can AI Predict AF Recurrence After Cardioversion? | JACC Baran
-
ナレーター：
-
著者：
Hosts Mitsuaki Sawano, MD, and Satoshi Shoji, MD welcome Dr. Keisuke Usuda, MD (Massachusetts General Hospital / Broad Institute of MIT and Harvard) to discuss his JACC: Clinical Electrophysiology study, "Electrocardiogram-Based Artificial Intelligence to Predict Recurrence After Cardioversion for Newly Diagnosed Atrial Fibrillation." Using a large electronic health record cohort of patients undergoing first-time direct current cardioversion (DCCV) for newly diagnosed atrial fibrillation, this study quantified long-term recurrence risk and evaluated whether an ECG-based AI model could improve prediction of AF recurrence. Over a median 4.4 years of follow-up, AF recurrence occurred in more than half of patients, with marked age-related differences in risk. Notably, an AI model derived from a single 12-lead ECG outperformed conventional clinical prediction models for recurrence discrimination. The episode explores how AI-driven ECG analysis may help personalize rhythm control strategies, identify patients who may benefit from intensified monitoring or early intervention, and reshape post-cardioversion management in atrial fibrillation.