Background. Azithromycin exposure has been reported to increase the risk of QT prolongation and cardiovascular death. However, findings on the association between azithromycin and cardiovascular death are controversial, and azithromycin is still used in actual practice. Additionally, quantitative assessments of risk have not been performed, including the risk of QT prolongation when patients are exposed to azithromycin in a real-world clinical setting. Therefore, in this study, we aimed to evaluate the risk of exposure to azithromycin on QT prolongation in a real-world clinical setting using a 21-year medical history database of a tertiary medical institution. Methods. We analyzed the electrocardiogram results and relevant electronic health records of 402,607 subjects in a tertiary teaching hospital in Korea from 1996 to 2015. To evaluate the risk of QT prolongation of azithromycin, we conducted a case-control analysis using amoxicillin for comparison. Multiple logistic regression analysis was performed to correct for age, sex, accompanying drugs, and disease. Results. The odds ratio (OR) for QT prolongation (QTc>450 ms in male and >460 ms in female) on azithromycin exposure was 1.40 (95% confidence interval [CI], 1.23-1.59), and the OR for severe QT prolongation (QTc>500 ms) was 1.43 (95% CI, 1.13-1.82). On the other hand, the ORs on exposure to amoxicillin were 1.06 (95% CI, 0.97-1.15) and 0.88 (95% CI, 0.70-1.09). In a subgroup analysis, the risk of QT prolongation in patients aged between 60 and 80 years was significantly higher when they are exposed to azithromycin. Conclusions. The risk of QT prolongation was increased when patients, particularly the elderly aged 60-79 years, were exposed to azithromycin. Therefore, clinicians should pay exercise caution using azithromycin or consider using other antibiotics, such as amoxicillin, instead of azithromycin.

Biosignal data captured by patient monitoring systems could provide key evidence for detecting or predicting critical clinical events; however, noise in these data hinders their use. Because deep learning algorithms can extract features without human annotation, this study hypothesized that they could be used to screen unacceptable electrocardiograms (ECGs) that include noise. To test that, a deep learning-based model for unacceptable ECG screening was developed, and its screening results were compared with the interpretations of a medical expert. To develop and apply the screening model, we used a biosignal database comprising 165,142,920 ECG II (10-second lead II electrocardiogram) data gathered between August 31, 2016 and September 30, 2018 from a trauma intensive-care unit. Then, 2,700 and 300 ECGs (ratio of 9:1) were reviewed by a medical expert and used for 9-fold cross-validation (training and validation) and test datasets. A convolutional neural network-based model for unacceptable ECG screening was developed based on the training and validation datasets. The model exhibiting the lowest cross-validation loss was subsequently selected as the final model. Its performance was evaluated through comparison with a test dataset. When the screening results of the proposed model were compared to the test dataset, the area under the receiver operating characteristic curve and the F1-score of the model were 0.93 and 0.80 (sensitivity = 0.88, specificity = 0.89, positive predictive value = 0.74, and negative predictive value = 0.96). The deep learning-based model developed in this study is capable of detecting and screening unacceptable ECGs efficiently.

X: original ECG data;

Cardiovascular disease (CVD) remains a pressing global health concern. While traditional risk prediction methods such as the Framingham and American College of Cardiology/American Heart Association (ACC/AHA) risk scores have been widely used in the practice, artificial intelligence (AI), especially GPT-4, offers new opportunities. Utilizing large scale of multi-center data from 47,468 UK Biobank participants and 5,718 KoGES participants, this study quantitatively evaluated the predictive capabilities of GPT-4 in comparison with traditional models. Our results suggest that the GPT-based score showed commendably comparable performance in CVD prediction when compared to traditional models (AUROC on UKB: 0.725 for GPT-4, 0.733 for ACC/AHA, 0.728 for Framingham; KoGES: 0.664 for GPT-4, 0.674 for ACC/AHA, 0.675 for Framingham). Even with omission of certain variables, GPT-4's performance was robust, demonstrating its adaptability to data-scarce situations. In conclusion, this study emphasizes the promising role of GPT-4 in predicting CVD risks across varied ethnic datasets, pointing toward its expansive future applications in the medical practice.

Most existing electrocardiogram (ECG) feature extraction methods rely on rule-based approaches. It is difficult to manually define all ECG features. We propose an unsupervised feature learning method using a convolutional variational autoencoder (CVAE) that can extract ECG features with unlabeled data. We used 596,000 ECG samples from 1,278 patients archived in biosignal databases from intensive care units to train the CVAE. Three external datasets were used for feature validation using two approaches. First, we explored the features without an additional training process. Clustering, latent space exploration, and anomaly detection were conducted. We confirmed that CVAE features reflected the various types of ECG rhythms. Second, we applied CVAE features to new tasks as input data and CVAE weights to weight initialization for different models for transfer learning for the classification of 12 types of arrhythmias. The f1-score for arrhythmia classification with extreme gradient boosting was 0.86 using CVAE features only. The f1-score of the model in which weights were initialized with the CVAE encoder was 5% better than that obtained with random initialization. Unsupervised feature learning with CVAE can extract the characteristics of various types of ECGs and can be an alternative to the feature extraction method for ECGs.