Performance of a convolutional neural network derived from an ECG database in recognizing myocardial infarction.

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.

Avoiding inappropriate therapy of single-lead implantable cardioverter-defibrillator by using atrial-sensing electrodes.

INTRODUCTION: The single-chamber implantable cardioverter-defibrillator (ICD) can be associated with more frequent inappropriate therapies compared with dual-chamber ICDs, when they are accompanied by a simpler implantation procedure. The aim of this study was to investigate whether the use of a single-lead ICD system with atrial-sensing electrodes results in a reduction of inappropriate ICD therapy. METHODS AND RESULTS: The study population consisted of 212 consecutive patients, who underwent primary prophylactic single-lead ICD implantation at our institute. A ventricular lead with atrial-sensing electrodes was implanted in 77 patients (36%; Group-VDD) and a ventricular lead without atrial-sensing electrodes was implanted in 135 patients (64%; Group-VVI). Procedural and follow-up data were collected in a prospective registry. A higher prevalence of atrial fibrillation was present in Group-VDD. There were no other significant differences in patient baseline characteristics (age, sex, and other comorbidities) or follow-up period between the two groups. The operative parameters including fluoroscopic burden showed no significant differences between Group-VDD and Group-VVI. During a mean follow-up period of 697 ± 392 days, 26 patients (12%) experienced appropriate ICD therapies and 13 patients (6%) suffered inappropriate ICD therapies. The incidence of inappropriate ICD therapies in Group-VDD was significantly lower as compared to that of Group-VVI (1/77 [1%] vs 12/135 [9%]; log-rank, P = 0.028). The incidence of appropriate ICD therapies and the occurrence of device-related complications showed no significant difference between the two groups. CONCLUSION: Single-lead ICD with atrial-sensing electrodes shows a lower incidence of inappropriate ICD therapy compared with the absence of atrial-sensing electrodes, without additional operative burden or increased complications.