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: Supraventricular Tachycardia; Atrioventricular node

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