Artificial intelligence–enabled electrocardiogram to distinguish atrioventricular re-entrant tachycardia from atrioventricular nodal re-entrant tachycardia

• DOI: 10.1016/j.cvdhj.2023.01.004
• Publication Date: 2023-01-31T03:02:57Z
• Authors: Arunashis Sau; Safi Ibrahim; Daniel B. Kramer; Jonathan W. Waks; Norman Qureshi; Michael Koa-Wing; Daniel Keene; Louisa Malcolme-L...; David C. Lefroy; Nicholas W.F. Linton; Phang Boon Lim; Amanda Varnava; Zachary I. Whinnett; Prapa Kanagaratnam; Danilo Mandic; Nicholas S. Peters; Fu Siong Ng
• Keywords: 03 medical and health sciences; 0302 clinical medicine; 610

Accurately determining arrhythmia mechanism from a 12-lead electrocardiogram (ECG) of supraventricular tachycardia can be challenging. We hypothesized a convolutional neural network (CNN) can be trained to classify atrioventricular re-entrant tachycardia (AVRT) vs atrioventricular nodal re-entrant tachycardia (AVNRT) from the 12-lead ECG, when using findings from the invasive electrophysiology (EP) study as the gold standard.We trained a CNN on data from 124 patients undergoing EP studies with a final diagnosis of AVRT or AVNRT. A total of 4962 5-second 12-lead ECG segments were used for training. Each case was labeled AVRT or AVNRT based on the findings of the EP study. The model performance was evaluated against a hold-out test set of 31 patients and compared to an existing manual algorithm.The model had an accuracy of 77.4% in distinguishing between AVRT and AVNRT. The area under the receiver operating characteristic curve was 0.80. In comparison, the existing manual algorithm achieved an accuracy of 67.7% on the same test set. Saliency mapping demonstrated the network used the expected sections of the ECGs for diagnoses; these were the QRS complexes that may contain retrograde P waves.We describe the first neural network trained to differentiate AVRT from AVNRT. Accurate diagnosis of arrhythmia mechanism from a 12-lead ECG could aid preprocedural counseling, consent, and procedure planning. The current accuracy from our neural network is modest but may be improved with a larger training dataset.