Artificial intelligence (AI) is developing rapidly in the medical technology field, particularly in image analysis. ECG-diagnosis is an image analysis in the sense that cardiologists assess the waveforms presented in a 2-dimensional image. We hypothesized that an AI using a convolutional neural network (CNN) may also recognize ECG images and patterns accurately. We used the PTB ECG database consisting of 289 ECGs including 148 myocardial infarction (MI) cases to develop a CNN to recognize MI in ECG. Our CNN model, equipped with 6-layer architecture, was trained with training-set ECGs. After that, our CNN and 10 physicians are tested with test-set ECGs and compared their MI recognition capability in metrics F1 (harmonic mean of precision and recall) and accuracy. The F1 and accuracy by our CNN were significantly higher (83 ± 4%, 81 ± 4%) as compared to physicians (70 ± 7%, 67 ± 7%, P < 0.0001, respectively). Furthermore, elimination of Goldberger-leads or ECG image compression up to quarter resolution did not significantly decrease the recognition capability. Deep learning with a simple CNN for image analysis may achieve a comparable capability to physicians in recognizing MI on ECG. Further investigation is warranted for the use of AI in ECG image assessment.

Electrophysiological studies in mice, the prevailing model organism in the field of basic cardiovascular research, are impeded by the low yield of programmed electrical stimulation (PES). To investigate a modified approach for ventricular arrhythmia (VA) induction and a novel scoring system in mice. A systematic review of literature on current methods for PES in mice searching the PubMed database revealed that VA inducibility was low and ranged widely (4.6 ± 10.7%). Based on this literature review, a modified PES protocol with 3 to 10 extrastimuli was developed and tested in comparison to the conventional PES protocol using up to 3 extrastimuli in anesthetized wildtype mice (C57BL/6J, n = 12). Induced VA, classified according to the Lambeth Convention, were assessed by established arrhythmia scores as well as a novel arrhythmia score based on VA duration. PES with the modified approach raised both the occurrence and the duration of VA compared to conventional PES (0% vs 50%; novel VA score p = 0.0002). Particularly, coupling of >6 extrastimuli raised the induction of VA. Predominantly, premature ventricular complexes (n = 6) and ventricular tachycardia <1s (n = 4) were observed. Repeated PES after adrenergic stimulation using isoprenaline resulted in enhanced induction of ventricular tachycardia <1s in both protocols. Our findings suggest that the presented approach of modified PES enables effective induction and quantification of VA in wildtype mice and may well be suited to document and evaluate detailed VA characteristics in mice.

The multipolar irrigated radiofrequency (RF) ablation catheter (nMARQ™) is a novel tool for pulmonary vein isolation (PVI). We investigated the incidence of thermal oesophageal injury (EI) using the nMARQ™ for PVI. In the initial six patients (Group 1), RF was delivered at the posterior wall with a maximum duration of 60 s and a maximum power (maxP) of 20 W for unipolar ablation, and a maxP of 10 W for the bipolar ablation. In the latter 15 patients (Group 2), RF application was limited at the posterior wall to a maximum duration of 30 s and a maxP of 15 W for unipolar ablation a max P of 10 W for bipolar ablation. Oesophageal temperature monitoring was performed in all patients and ablation was terminated at a temperature rise >41°C. Endoscopy was carried out within 2 days post-ablation. Pulmonary vein isolation was performed during sinus rhythm and was successfully achieved in 83 of 84 PVs except the septal inferior vein in one patient. Charring was seen in 3 of 21 (14.3%) patients without any evidence of embolism. Phrenic nerve palsy occurred in one patient. Endoscopy revealed severe EI in 3 of 6 (50%) patients in Group 1 and in 1 of 15 patients (6.7%) in Group 2. Procedure times between Groups 1 and 2 were similar (228.3 ± 60.2 min vs. 221.3 ± 51.8 min; P = 0.79). An unexpectedly high incidence of thermal EI was noted following PVI using the nMARQ™ with the initial ablation protocol. However, the incidence of thermal EI can be sigificantly reduced with limited power and RF application time at the posterior left atrium.