Hellenic Journal of Cardiology最新文献

Objective: Electrocardiography (ECGs) has been a vital tool for cardiovascular disease (CVD) diagnosis, which visually depicts the heart's electrical activity. To enhance automatic classification between normal and diseased ECG, it is essential to extract consistent and qualitative features.

Methods: Precision of ECG classification through a hybrid Deep Learning (DL) approach leverages both Convolutional Neural Network (CNN) architecture and Variational Autoencoder (VAE) techniques. By combining these methods, we aim to achieve more accurate and robust ECG interpretation. The method is trained and tested over the PTB-XL dataset, which contains 21,799 with 12-lead ECGs from 18,869 patients, each spanning 10 s. The classification evaluation of five super-classes and 23 sub-classes of CVD, with the proposed CNN-VAE model is compared.

Results: The classification of various CVDs resulted in the highest accuracy of 98.51%, specificity of 98.12%, sensitivity of 97.9%, and F1-score of 97.95%. We have also achieved the minimum false positive and false negative rates of 2.07% and 1.87%, respectively, during validation. The results are validated upon the annotations given by individual cardiologists, who assigned potentially multiple ECG statements to each record.

Conclusion: When compared to other deep learning methods, our suggested CNN-VAE model performs significantly better in the testing phase. This study proposes a new architecture of combining CNN-VAE for CVD classification from ECG data, this can help clinicians to identify the disease earlier and carry out further treatment. The CNN-VAE model can better characterize input signals due to its hybrid architecture.

Objective: This nationwide study aims to analyze mortality trends for all individual causes in Greece from 2001 to 2020, with a specific focus on 2020, a year influenced by the COVID-19 pandemic. As Greece is the fastest-aging country in Europe, the study's findings can be generalized to other aging societies, guiding the reevaluation of global health policies.

Methods: Data on the population and the number of deaths were retrieved from the Hellenic Statistical Authority. We calculated age-standardized mortality rates (ASMR) and cause-specific mortality rates by sex in three age groups (0-64, 65-79, and 80+ years) from 2001 to 2020. Proportional mortality rates for 2020 were determined. Statistical analysis used generalized linear models with Python Programming Language.

Results: From 2001 to 2020, the ASMR of cardiovascular diseases (CVD) decreased by 42.7% (p < 0.0001), with declines in most sub-causes, except for hypertensive diseases, which increased by 2.8-fold (p < 0.0001). In 2020, the proportional mortality rates of the three leading causes were 34.9% for CVD, 23.5% for neoplasms, and 9.6% for respiratory diseases (RD). In 2020, CVD were the leading cause of death among individuals aged 80+ years (39.3%), while neoplasms were the leading cause among those aged 0-79 years (37.7%). Among cardiovascular sub-causes, cerebrovascular diseases were predominant in the 80+ year age group (30.3%), while ischemic heart diseases were most prevalent among those aged 0-79 years (up to 60.0%).

Conclusions: The global phenomenon of population aging necessitates a reframing of health policies in our aging societies, focusing on diseases with either a high mortality burden, such as CVD, neoplasms, and RD, or those experiencing increasing trends, such as hypertensive diseases.

Background: We assessed left ventricular ejection fraction (LVEF) to compare the effects of renin-angiotensin system inhibitors (RASI) in patients with non-ST-segment elevation myocardial infarction (NSTEMI).

Methods: We categorized 4558 patients with NSTEMI as either RASI users (3752 patients) or non-users (806 patients). The 3-year patient-oriented composite outcome (POCO), which included all-cause death, recurrent myocardial infarction, any repeat revascularization, or hospitalization for heart failure (HF), was the primary outcome. To compare clinical outcomes, a multivariable-adjusted hazard ratio (aHR) was calculated after performing multicollinearity tests on all significant confounding variables (P < 0.05).

Results: Among RASI users, the aHRs for POCO, all-cause death, and cardiac death were significantly higher in the HF with reduced EF (HFrEF) subgroup than in the HF with mildly reduced EF (HFmrEF) (1.610, 2.120, and 2.489; P < 0.001, <0.001, and <0.001; respectively) and HF with preserved EF (HFpEF) (2.234, 3.920, and 5.215; P < 0.001, <0.001, and <0.001; respectively) subgroups. The aHRs for these variables were significantly higher in the HFmrEF subgroup than the HFpEF subgroup (1.416, 1.843, and 2.172, respectively). Among RASI non-users, the aHRs for these variables were significantly higher in the HFrEF subgroup than the HFmrEF (2.573, 3.172, and 3.762, respectively) and HFpEF (2.425, 3.805, and 4.178, respectively) subgroups. In three LVEF subgroups, RASI users exhibited lower aHRs for POCO and all-cause death than RASI non-users.

Conclusion: In the RASI users group, the aHRs for POCO and mortality were highest in the HFrEF subgroup, intermediate in the HFmrEF subgroup, and lowest in the HFpEF subgroup.

Objective: The benefits of rhythm control in early atrial fibrillation (AF) are increasingly recognized. This study aimed to investigate whether early AF ablation contributes to long-term sinus rhythm maintenance and to identify a suitable predictive score.

Methods: According to diagnosis-to-ablation time, this study prospectively enrolled 245 patients with very early AF, 262 with early AF, and 588 with late AF for radiofrequency ablation from June 2017 to December 2022. Clinical data, risk scores, and follow-up results were collected and analyzed.

Results: Baseline characteristics were similar among the three cohorts. During a median follow-up period of 26 months, AF recurrence was observed in 61 (24.9%), 66 (25.2%), and 216 (36.7%) patients in the very early, early, and late AF cohorts, respectively. In the multivariable-adjusted model, very early and early AF were associated with a reduced risk of AF recurrence, with hazard ratios of 0.72 (95% confidence interval [CI] 0.52-0.99) and 0.57 (95% CI 0.41-0.78), respectively. The APPLE score demonstrated the highest predictive power for very early AF, with an area under the curve (AUC) of 0.74. However, its predictive power decreased with time from diagnosis, showing low predictive power for late AF (AUC = 0.58). In addition, the time-dependent concordance index showed consistent results. For very early AF, the Akaike information criterion and decision curve analysis showed that APPLE had the highest predictive value.

Conclusion: Very early AF ablation was associated with a lower recurrence rate, and the APPLE score provided a higher predictive value for these patients. (URL: https://www.chictr.org.cn/; Unique identifier: ChiCTR-OIN-17013021).

Objective: This study aimed to leverage real-world electronic medical record data to develop interpretable machine learning models for diagnosis of Kawasaki disease while also exploring and prioritizing the significant risk factors.

Methods: A comprehensive study was conducted on 4087 pediatric patients at the Children's Hospital of Chongqing, China. The study collected demographic data, physical examination results, and laboratory findings. Statistical analyses were performed using IBM SPSS Statistics, Version 26.0. The optimal feature subset was used to develop intelligent diagnostic prediction models based on the Light Gradient Boosting Machine, Explainable Boosting Machine (EBM), Gradient Boosting Classifier (GBC), Fast Interpretable Greedy-Tree Sums, Decision Tree, AdaBoost Classifier, and Logistic Regression. Model performance was evaluated in three dimensions: discriminative ability via receiver operating characteristic curves, calibration accuracy using calibration curves, and interpretability through SHAP (SHapley Additive exPlanations) and LIME (Local Interpretable Model-Agnostic Explanations).

Results: In this study, Kawasaki disease was diagnosed in 2971 participants. Analysis was conducted on 31 indicators, including red blood cell distribution width and erythrocyte sedimentation rate. The EBM model demonstrated superior performance relative to other models, with an area under the curve of 0.97, second only to the GBC model. Furthermore, the EBM model exhibited the highest calibration accuracy and maintained its interpretability without relying on external analytical tools such as SHAP and LIME, thus reducing interpretation biases. Platelet distribution width, total protein, and erythrocyte sedimentation rate were identified by the model as significant predictors for the diagnosis of Kawasaki disease.

Conclusion: This study used diverse machine learning models for early diagnosis of Kawasaki disease. The findings demonstrated that interpretable models such as EBM outperformed traditional machine learning models in terms of both interpretability and performance. Ensuring consistency between predictive models and clinical evidence is crucial for the successful integration of artificial intelligence into real-world clinical practice.

Objective: Aortic dissection remains a life-threatening condition necessitating accurate diagnosis and timely intervention. This study aimed to investigate phenotypic heterogeneity in patients with Stanford type B aortic dissection (TBAD) through machine learning clustering analysis of cardiovascular computed tomography (CT) imaging.

Methods: Electronic medical records were collected to extract demographic and clinical features of patients with TBAD. Exclusion criteria ensured homogeneity and clinical relevance of the TBAD cohort. Controls were selected on the basis of age, comorbidity status, and imaging availability. Aortic morphological parameters were extracted from CT angiography and subjected to K-means clustering analysis to identify distinct phenotypes.

Results: Clustering analysis revealed three phenotypes of patients with TBAD with significant correlations with population characteristics and dissection rates. This pioneering study used CT-based three-dimensional reconstruction to classify high-risk individuals, demonstrating the potential of machine learning in enhancing diagnostic accuracy and personalized treatment strategies. Recent advancements in machine learning have garnered attention in cardiovascular imaging, particularly in aortic dissection research. These studies leverage various imaging modalities to extract valuable features and information from cardiovascular images, paving the way for more personalized interventions.

Conclusion: This study provides insights into the phenotypic heterogeneity of patients with TBAD using machine learning clustering analysis of cardiovascular CT imaging. The identified phenotypes exhibit correlations with population characteristics and dissection rates, highlighting the potential of machine learning in risk stratification and personalized management of aortic dissection. Further research in this field holds promise for improving diagnostic accuracy and treatment outcomes in patients with aortic dissection.

Objective: The echocardiographic assessment of the systemic right ventricle (sRV) performance during stress testing is limited and evaluation is not routinely performed. The aim of the study is to investigate sRV myocardial performance at rest and with exercise in patients with complete transposition of the great arteries (dTGA) who have undergone atrial switch operation.

Methods: In a single-center cross-sectional study, 41 patients with dTGA following the atrial switch operation and gender-matched 20 healthy volunteers underwent exercise echocardiography on a bicycle ergometer in the semi-supine position to assess sRV systolic function indices: tricuspid annular plane systolic excursion (TAPSE), right ventricular area change (FAC), global longitudinal strain (GLS) and systemic velocity time integral (VTI).

Results: Patients with sRV were characterized by lower systolic function assessed by TAPSE, s', FAC, GLS both at baseline and at peak exercise, compared with the control group. sRV GLS decreased during exercise in patients with sRV (-6 + 2.84) compared to increased in patients with systemic left ventricle (0.47 + 2.74), p < 0.001. There was no increase in VTI during exercise in patients with sRV, compared to controls (Δ VTI -0.01 ± 2.96 cm vs. Δ VTI 4.50 ± 3.13 cm, p < 0.001). There was a trend towards higher chronotropic incompetence in patients with sRV vs. control (61% vs. 45%, p = 0.28).

Conclusion: Our results confirmed that patients with dTGA have reduced ability to increase myocardial contractility and stroke volume during exercise. Chronotropic incompetence was prevalent in dTGA patients.