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.

Major adverse cardiac events and drug-coated balloon size in coronary interventions.

OBJECTIVE: In-stent restenosis (ISR) is a feared complication after coronary stent implantation. Drug-coated balloon (DCB) is being promoted as a treatment option for ISR. However, the benefit-risk ratio of DCB length has not been investigated. Longer DCBs release more anti-proliferative drug to the vessel wall; however, they are associated with a higher lesion length and vessel injury. HYPOTHESIS: DCB length is associated with clinical outcome. METHODS: We analyzed 286 consecutive Pantera Lux (Biotronik, active component Paclitaxel) DCB-treated patients between April 2009 and June 2012. Of them, 176 patients were treated using a 15-mm DCB and 109 were treated using a 20-mm DCB. Baseline characteristics and major adverse cardiac events (MACE; death, myocardial infarction, and target lesion revascularization) during initial hospital stay and a 2-year follow-up period were obtained. RESULTS: Patients characteristics such as cardiovascular risk factors, prior diseases, co-medication, clinical presentation, target vessel, and left ventricular function did not differ between the groups. MACE during hospital course was similar [1.7% vs. 2.8%, relative risk (RR) 1.6, 95% confidence interval (CI) 0.3-7.9, p=0.554]. Likewise, at 2-year follow-up, MACE did not differ between the groups (23.2% vs. 27.5%, RR 1.2, 95% CI 0.6-1.5, p=0.408). CONCLUSION: DCB length was not associated with clinical outcome during a 2-year follow-up period.

SeQuent Please vs. Pantera Lux drug coated balloon angioplasty in real life: Results from the Düsseldorf DCB registry.

BACKGROUND: In-stent restenosis (ISR) is still a major concern in interventional cardiology. Drug coated balloon (DCB) angioplasty has been shown to be a promising option in treatment of ISR. However heterogeneity of different DCBs in suppression of neointimal growth has been described in a porcine model of coronary ISR. Therefore, in this registry analysis, we compared two frequently used paclitaxel eluting DCBs, the SeQuent Please and the Pantera Lux DCB. METHODS: 571 patients were treated with DCB angioplasty at the Heinrich-Heine University Düsseldorf between 2009 and 2012. Follow-up was conducted during ambulatory care at our department. Major adverse cardiac events (death, myocardial infarction [MI] and target lesion revascularization) were registered during hospitalization and follow-up. RESULTS: Patient characteristics, prior diseases, clinical presentation, ejection fraction, procedural success and lost-for-follow-up did not differ between patients treated with the SeQuent Please and. The Pantera Lux DCB. MACE during hospital course were similar as well (Pantera Lux: 6 patients [1.6%] vs. SeQuent®Please: 3 patients [1.5%], relative risk 1.06, 95% confidence interval 0.3-4.2, P=0.93). Event free survival was significantly longer in patients treated with the Pantera Lux DCB as compared to SeQuent Please DCB (Hazard ratio: 0.65, 95% confidence interval 0.43-0.98; P value of log-rank test: 0.0405). CONCLUSION: MACE free survival was longer in Pantera Lux DCB treated patients as compared to SeQuent Please treated patients. This finding has to be confirmed in future clinical trials.