Tolga Çakmak

Background/Objectives: Cardiovascular diseases remain a major global health burden, while echocardiography, despite its diagnostic value, is not always readily accessible. This study proposes EchoGraph-FusionNet, a multimodal deep learning framework for predicting echocardiographic abnormalities and structural heart disease (SHD) from 12-lead electrocardiograms (ECGs) and clinical variables. Methods: The framework integrates ECG and clinical representations through cross-modal fusion, followed by a disease-aware mixture-of-experts module and a two-layer adaptive graph attention network that jointly models 11 echocardiographic abnormalities. Five continuous echocardiographic measurements were incorporated as auxiliary regression tasks. Prevalence-aware logit adjustment and asymmetric loss were employed to address class imbalance. Performance was evaluated on the predefined EchoNext training, validation, and held-out test partitions across three independent random seeds. Results: On the held-out test set, EchoGraph-FusionNet achieved a macro AUROC of 0.8066 ± 0.0025 and macro AUPRC of 0.2838 ± 0.0056 across 11 abnormalities. For SHD detection, AUROC and AUPRC reached 0.8345 ± 0.0001 and 0.8039 ± 0.0031, respectively. Conclusions: EchoGraph-FusionNet effectively combines multimodal representations, disease relationships, and multitask learning for ECG-based prediction of echocardiographic abnormalities, supporting its potential as a screening and prioritization approach for structural cardiac assessment.

Background and Objectives: Although chronic total occlusion (CTO) often presents with varying degrees of coronary collateral circulation (CCC), the biological factors accompanying inadequate collateralization are not fully established. The Remnant Cholesterol Inflammation Index (RCII) is a composite metric integrating atherogenic lipid load and systemic inflammation. This study examined the association between RCII and CCC and the discriminative ability of RCII for identifying insufficient collateral networks in patients with CTO. Materials and Methods: This retrospective, single-center analysis evaluated 359 patients with at least one major epicardial CTO who underwent diagnostic coronary angiography between 15 January 2023 and 15 January 2026. Participants were stratified via the Rentrop scoring system into poor (grades 0–1; n = 168) and good (grades 2–3; n = 191) CCC cohorts. RCII was calculated as [remnant cholesterol (mg/dL) × high-sensitivity C-reactive protein (hs-CRP, mg/L)]/10, with remnant cholesterol derived as total cholesterol minus LDL-C and HDL-C. Univariable and multivariable logistic regression models, together with receiver operating characteristic (ROC) curve analysis, were used to assess the association of RCII with poor CCC and its discriminative ability. Results: Patients with poor CCC exhibited significantly higher RCII values compared with the good CCC cohort (19.7 ± 4.7 vs. 6.5 ± 2.4, p < 0.001). Following multivariable adjustment, higher RCII remained independently associated with poor collateralization (OR 1.120, 95% CI 1.052–1.199; p < 0.001), alongside right coronary artery (RCA) CTO and the systemic immune-inflammation index (SII). RCII showed discriminative ability for poor CCC (AUC 0.792, 95% CI 0.745–0.835; p < 0.001), which was greater than that of remnant cholesterol, hs-CRP, or SII considered separately. In this cohort, an RCII threshold above 12.3 identified poor CCC with 80% sensitivity and 71% specificity. Conclusions: In this retrospective cohort, higher RCII was independently associated with inadequate coronary collateralization in the setting of CTO. RCII may represent a candidate marker of the combined residual lipid and inflammatory burden; prospective studies and external validation, together with formal assessment of clinical utility, are required before any clinical application can be considered.

Background/Objectives: Cardiovascular diseases remain a leading cause of death worldwide, and early diagnosis is essential. Phonocardiogram (PCG) signals provide a non-invasive and inexpensive means of cardiac assessment. However, accurate interpretation of PCG signals requires clinical expertise. In this paper, we propose TriFusion-PCGNet, a hybrid multi-branch deep learning framework for automated heart sound classification. Methods: The proposed method was evaluated on the PhysioNet/Computing in Cardiology Challenge 2016 dataset. Three complementary PCG representations—the raw signal, band-pass filtered signal, and Hilbert envelope signal—were processed through dedicated feature extraction branches using residual convolutional networks, dilated residual convolutions, and a BiGRU with a temporal attention mechanism. The extracted features were fused and classified using fully connected layers. Performance was evaluated using the original train–validation split and stratified 10-fold cross-validation. Results: Under a strictly separated record-level 10-fold cross-validation protocol, TriFusion-PCGNet achieved 90.87% mean accuracy and 95.34% AUC, with outer test folds isolated from training and model selection. Subject-level analysis showed substantially lower performance, emphasizing the importance of subject-disjoint evaluation for patient-level generalization. Conclusions: TriFusion-PCGNet integrates complementary PCG representations for automated normal/abnormal heart sound classification. However, the substantially lower subject-disjoint performance indicates that the record-level results should not be interpreted as evidence of patient-independent generalization, and further validation on independent subject-annotated datasets is required.

Background/Objectives: Cardiovascular diseases are among the leading causes of death worldwide. Early diagnosis of these conditions minimizes the risk of future death. Listening to heart sounds with a stethoscope is one of the easiest and fastest methods for diagnosing heart conditions. While heart sounds are a quick and easy diagnostic method, they require significant expert interpretation. Recently, artificial intelligence models trained based on these expert interpretations have become popular in the development of decision support systems. Methods: The proposed approach uses the popular 2016 PhysioNet/CinC Challenge dataset for PCG signals. Spectrogram image transformation was then performed to increase the representativeness of these signals. A deep learning-based model that allows for the simultaneous training of residual and attention blocks and the MLP-mixer model was used for feature extraction. A new algorithm combining the strengths of NCA and ReliefF algorithms was proposed to select the strongest features in the feature set. The SVM algorithm was used for classification. Results: With this proposed approach, over 98% success was achieved in all accuracy, sensitivity, specificity, precision, and F1-score metrics. Conclusions: As a result, an artificial intelligence-based decision support system that detects cardiovascular diseases with high accuracy is presented.

Background/Objectives: Heart attacks are the leading cause of death in the world. There are two important classes of heart attack: ST-segment Elevation Myocardial Infarction (STEMI) and Non-ST-segment Elevation Myocardial Infarction (NSTEMI) patient groups. While the STEMI group has a higher mortality rate in the short term, the NSTEMI group is considered more dangerous and insidious in the long term. Blocked coronary arteries can be predicted from ECG signals in STEMI patients but not in NSTEMI patients. Therefore, coronary angiography (CAG) is inevitable for these patients. However, in the elderly and some patients with chronic diseases, if there is a single blockage, the CAG procedure poses a risk, so medication may be preferred. In this study, a novel deep learning-based approach is used to automatically detect the occluded main coronary artery or arteries in NSTEMI patients. For this purpose, a new seven-class dataset was created with expert cardiologists. Methods: A new Multi Input-Multi Scale (MI-MS) ConvMixer model was developed for automatic detection. The MI-MS ConvMixer model allows simultaneous training of 12-channel ECG data and highlights different regions of the data at different scales. In addition, the ConMixer structure provides high classification performance without increasing the complexity of the model. Moreover, to maximise the classifier performance, the WSSE algorithm was developed to adjust the classification prediction value according to the feature importance weights. Results: This algorithm improves the SVM classifier performance. The features extracted from this model were classified with the WSSE algorithm, and an accuracy of 88.72% was achieved. Conclusions: This study demonstrates the potential of the MI-MS ConvMixer model in advancing ECG signal classification for diagnosing coronary artery diseases, offering a promising tool for real-time, automated analysis in clinical settings. The findings highlight the model’s ability to achieve high sensitivity, specificity, and precision, which could significantly improve.

Aims: The present study explored if pentraxin-3 (PTX-3) levels, which would be boosted due to cardiac damage by gentamicin, can be regressed thanks to Pycnogenol, which was also previously shown to have desirable impacts on cardiovascular diseases. Methods: In the study, we recruited 28 8-10-week-old male Sprague-Dawley rats into four groups: control, gentamicin, gentamycin+Pycnogenol, and Pycnogenol. We stained the tissue samples with hematoxylin-eosin and Masson’s trichome dye for histopathological analysis. Then, malondialdehyde (MDA) levels were measured using the spectrophotometric technique. In addition, we measured PTX-3 levels in the heart tissues by an immunohistochemical method. Results: We discovered the heart tissue samples of the rats in the control and Pycnogenol groups were histologically normal. As well as mononuclear cell increase and degeneration of cardiac muscle cells, we observed mild congestion in the gentamicin group compared to the control group. Despite more significant damage to the heart tissue in the gentamicin+Pycnogenol group compared to the control group, we found that the histopathological damage regressed in this group compared to the gentamicin group. While PTX-3 immunoreactivity was similar between the control and Pycnogenol groups, it was significantly elevated in the gentamicin group compared to the control group (p < 0.001). Moreover, the gentamicin+Pycnogenol group had decreased PTX-3 immunoreactivity than the gentamicin group. While MDA values followed a similar pattern between the control and Pycnogenol groups, these values were found to be significantly increased in the gentamicin group compared to the control group (p < 0.001). These values, however, were decreased in the gentamicin+Pycnogenol group compared to the gentamicin group. Conclusion: In a nutshell, the present study was able to demonstrate that gentamicin may lead to cardiac damage by boosting PTX-3 levels and that the damage can be regressed thanks to the Pycnogenol treatment.

Aims: Previous studies have shown association between hypertension and osteoporosis, however has not been evaluated alterations of circadian blood pressure patterns. The present study investigated the effects of osteoporosison the circadian blood pressure assessed by ambulatory blood pressure monitoring. Methods: 68 patients (mean age: 61.3±8.7 years, 34 dipper and 34 non-dipper patients) with postmenopausal hypertension (HT) were prospectively enrolled in this study. Ambulatory blood pressure monitoring was performed on all patients. We measured bone mineral density (BMD) and T-score by dual X- ray absorptiometry at the lumbar spine. The receiver operating characteristics (ROC) curve was used to demonstrate the sensitivity and specificity of lumbar spine BMD, optimal cut - off value for predicting non-dipper hypertension (NDHT). Results: T-score and BMD in non-dipper patients were lower than the dipper patients. There was a significant correlation between the rate of fall in night of mean blood pressure and both lumbar spine T-score and lumbar spine BMD (r= 0.330 and P =0.006 vs r=0.322 and P =0.007, respectively). A lumbar spine BMD <0.944 g/cm2 measured by DXA had a 64.7% sensitivity and 64.7% specificity in predicting non-dipper hypertension at ROC curve analysis. Conclusion: Low lumbar spine BMD is an independent and strong predictor of non-dipper hypertension. Osteoporosis may be an indicator of cardiovascular risk. Therefore, osteoporotic patients should be treated more aggressively.

Resumo Fundamentos Os efeitos protetores da fase de leitura aberta mitocondrial do 12S rRNA-c (MOTS-C) em doenças cardiovasculares foram demonstrados em vários estudos. Entretanto, há pouca documentação da relação entre MOTS-C e fluxo sanguíneo coronariano no infarto do miocárdio com supradesnivelamento do segmento ST (IAMCSST). Objetivo Nosso objetivo foi investigar o papel do MOTS-C, que é conhecido por ter propriedades citoprotetoras na patogênese do fenômeno de no-reflow, comparando a taxa de fluxo coronariano e os níveis de MOTS-C em pacientes [...] Resumo Fundamentos Os efeitos protetores da fase de leitura aberta mitocondrial do 12S rRNA-c (MOTS-C) em doenças cardiovasculares foram demonstrados em vários estudos. Entretanto, há pouca documentação da relação entre MOTS-C e fluxo sanguíneo coronariano no infarto do miocárdio com supradesnivelamento do segmento ST (IAMCSST). Objetivo Nosso objetivo foi investigar o papel do MOTS-C, que é conhecido por ter propriedades citoprotetoras na patogênese do fenômeno de no-reflow, comparando a taxa de fluxo coronariano e os níveis de MOTS-C em pacientes