Search Results (247)

Background and Clinical Significance: Cardiac sarcoidosis (CS) is a rare but life-threatening disorder, occurring in 2–5% of sarcoidosis cases, though post-mortem studies suggest a higher prevalence. It presents diagnostic challenges due to nonspecific symptoms and the low sensitivity of an endomyocardial biopsy. Recent guidelines emphasize multimodal imaging, such as cardiac magnetic resonance imaging (MRI) and positron emission tomography (PET). Given the risk of heart failure (HF) and arrhythmias, early detection is critical. This case highlights the role of non-invasive imaging in diagnosing CS and guiding treatment. Case Presentation: A 54-year-old female with asthma, hyperlipidemia, a recent diagnosis of anterior uveitis, and familial sarcoidosis presented with dyspnea, chest tightness, and worsening cough. Examination revealed anterior uveitis, erythema nodosum, jugular venous distension, and pedal edema. The electrocardiogram (ECG) demonstrated bifascicular block and premature ventricular contractions (PVCs). The brain natriuretic peptide (BNP) was 975 pg/mL, with the transthoracic echocardiogram revealing a left ventricular ejection fraction of 25–30% with global LV akinesis. Coronary computed tomography angiography (CCTA) excluded coronary artery disease. Cardiac MRI showed late gadolinium enhancement, with PET demonstrating active myocardial inflammation, supporting a >90% probability of CS. Given her clinical trajectory and risk of further decompensation, immunosuppressive therapy was initiated without pursuing a biopsy. A dual-chamber implantable cardioverter defibrillator (ICD) was placed due to risk of ventricular arrhythmias. Bronchoalveolar lavage (BAL) showed a CD4/CD8 ratio of 6.53, reinforcing the diagnosis. She responded well to treatment, with symptom improvement and repeat imaging demonstrating signs of disease remission. Conclusions: This case underscores the growing role of multimodal imaging in CS diagnosis, potentially replacing biopsy in select cases. Early imaging-based diagnosis enabled timely immunosuppression and ICD placement, improving outcomes. Full article

COVID-19 can cause acute infectious diseases of the respiratory system, and may probably lead to heart damage, which will seriously threaten human health. Electrocardiograms (ECGs) have the advantages of being low cost, non-invasive, and radiation free, and is widely used for evaluating heart health status. In this work, a lightweight deep learning network named GM-CBAM-ResNet is proposed for diagnosing COVID-19 based on ECG images. GM-CBAM-ResNet is constructed by replacing the convolution module with the Ghost module (GM) and adding the convolutional block attention module (CBAM) in the residual module of ResNet. To reveal the superiority of GM-CBAM-ResNet, the other three methods (ResNet, GM-ResNet, and CBAM-ResNet) are also analyzed from the following aspects: model performance, complexity, and interpretability. The model performance is evaluated by using the open ‘ECG Images dataset of Cardiac and COVID-19 Patients’. The complexity is reflected by comparing the number of model parameters. The interpretability is analyzed by utilizing Gradient-weighted Class Activation Mapping (Grad-CAM). Parameter statistics indicate that, on the basis of ResNet19, the number of model parameters of GM-CBAM-ResNet19 is reduced by 45.4%. Experimental results show that, under less model complexity, GM-CBAM-ResNet19 improves the diagnostic accuracy by approximately 5% in comparison with ResNet19. Additionally, the interpretability analysis shows that CBAM can suppress the interference of grid backgrounds and ensure higher diagnostic accuracy under lower model complexity. This work provides a lightweight solution for the rapid and accurate diagnosing of COVD-19 based on ECG images, which holds significant practical deployment value. Full article

Neuromuscular diseases (NMDs) encompass various hereditary conditions affecting motor neurons, the neuromuscular junction, and skeletal muscles. These disorders are characterized by progressive muscle weakness and can manifest at different stages of life, from birth to adulthood. NMDs, such as Duchenne and Becker muscular dystrophies, myotonic dystrophy, and limb–girdle muscular dystrophies, often involve cardiac complications, including cardiomyopathies and arrhythmias. Underlying genetic mutations contribute to skeletal and cardiac muscle dysfunction, particularly in the DMD, EMD, and LMNA genes. The progressive nature of muscle deterioration significantly reduces life expectancy, mainly due to respiratory and cardiac failure. The early detection of cardiac involvement through electrocardiography (ECG) and cardiac imaging is crucial for timely intervention. Pharmacological treatment focuses on managing cardiomyopathies and arrhythmias, with an emerging interest in gene therapies aimed at correcting underlying genetic defects. Heart transplantation, though historically controversial in patients with muscular dystrophies, is increasingly recognized as a viable option for individuals with advanced heart failure and moderate muscle impairment, leading to improved survival rates. Careful patient selection and management are critical to optimizing outcomes in these complex cases. Full article

There is no established detecting tool for hypertrophic cardiomyopathy (HCM) and dilated cardiomyopathy (DCM). This study aimed to develop a deep-learning-based model for identifying HCM and DCM using standard 12-lead electrocardiogram (ECG) images. We obtained a cohort of patients with HCM (171 ECG images) or DCM (364 ECG images), confirmed by cardiovascular magnetic resonance (CMR) examinations, who underwent both ECG and CMR within 30 days at our institution. Age- and sex-matched healthy controls (2314 ECG images) were selected from our Health Check Center. A total of 2849 ECG images were processed via a fine-tuned ResNet50 architecture, with stratified five-fold cross-validation for model training, validation, and testing. The proposed model demonstrated strong performance in distinguishing DCM, achieving an area under the receiver operating curve (AUROC) of 0.996 and an area under the precision–recall curve (AUPRC) of 0.940. For the detection of HCM, the model also achieved an AUROC of 0.980 and an AUPRC of 0.953, respectively. The model prospectively exhibited stability in temporal validation. Furthermore, representative images of the Gradient-weighted Class Activation Mapping (Grad-CAM) technique analysis showed the regions corresponding to the anterior and anteroseptal leads were the most important areas for the prediction of HCM or DCM. This temporally validated fine-tuned ResNet50 model shows promise to inexpensively detect individuals with HCM or DCM. Full article

Our research introduces a framework that integrates edge computing, quantum transfer learning, and federated learning to revolutionize pain level assessment through ECG signal analysis. The primary focus lies in developing a robust, privacy-preserving system that accurately classifies pain levels (low, medium, and high) by leveraging the intricate relationship between pain perception and autonomic nervous system responses captured in ECG signals. At the heart of our methodology lies a signal processing approach that transforms one-dimensional ECG signals into rich, two-dimensional Continuous Wavelet Transform (CWT) images. These transformations capture both temporal and frequency characteristics of pain-induced cardiac variations, providing a comprehensive representation of autonomic nervous system responses to different pain intensities. Our framework processes these CWT images through a sophisticated quantum–classical hybrid architecture, where edge devices perform initial preprocessing and feature extraction while maintaining data privacy. The cornerstone of our system is a Quantum Convolutional Hybrid Neural Network (QCHNN) that harnesses quantum entanglement properties to enhance feature detection and classification robustness. This quantum-enhanced approach is seamlessly integrated into a federated learning framework, allowing distributed training across multiple healthcare facilities while preserving patient privacy through secure aggregation protocols. The QCHNN demonstrated remarkable performance, achieving a classification accuracy of 94.8% in pain level assessment, significantly outperforming traditional machine learning approaches. Full article

Stress is a natural response of the organism to challenging situations, but its accurate detection is challenging due to its subjective nature. This study proposes a model based on depth-separable convolutional neural networks (DSCNN) to analyze heart rate variability (HRV) and detect stress. Electrocardiogram (ECG) signals are pre-processed to remove noise and ensure data quality. The signals are then transformed into two-dimensional images using the continuous wavelet transform (CWT) to identify pattern recognition in the time–frequency domain. These representations are classified using the DSCNN model to determine the presence of stress. The methodology has been validated using the SWELL-KW dataset, achieving an accuracy of 99.9% by analyzing the variability in three states (neutral, time pressure, and interruptions) of the 25 samples in the experiment, scanning the acquired signal every 5 s for 45 min per state. The proposed approach is characterized by its ability to transform ECG signals into time–frequency representations by means of short duration sampling, achieving an accurate classification of stress states without the need for complex feature extraction processes. This model is an efficient and accurate tool for stress analysis from biomedical signals. Full article

This paper seeks to solve the classification problem of cardiac arrhythmias by using a small number of electrocardiogram (ECG) recordings. To offer a reasonable solution to this problem, a technique that combines a common matrix approach (CMA)-based classifier model with the Fourier decomposition method (FDM) is proposed. The FDM is responsible for generating time–frequency (T-F) representations of ECG recordings. The classification process is performed with feature images applied as input to the classifier model. The feature images are obtained after two-dimensional principal component analysis (2DPCA) of data matrices related to ECG recordings. Each data matrix is created by concatenating the ECG record itself, the Fourier transform, and the T-F representation on a single matrix. To verify the efficacy of the proposed method, various experiments are conducted with the MIT-BIH, Chapman, and PTB-XL databases. In the assessments using the MIT-BIH database under the inter-patient paradigm, we achieved a mean overall accuracy rate of 99.81%. The proposed method outperforms the majority of recent efforts, yielding rates exceeding 99% on nearly five performance metrics for the recognition of V- and S-class arrhythmias. It is found that, in the classification of four types of arrhythmias using ECG recordings from the Chapman database, our model surpasses recent works by reaching mean overall accuracy rates of 99.76% and 99.45% for the raw and de-noised ECG recordings, respectively. Similarly, five different forms of arrhythmias from the PTB-XL database were recognized with a mean overall accuracy of 98.71%. Full article

Brugada syndrome (BrS) is an inherited electrical cardiac disorder that is associated with a higher risk of ventricular fibrillation (VF) and sudden cardiac death (SCD) in patients without structural heart disease. The diagnosis is based on the documentation of the typical pattern in the electrocardiogram (ECG) characterized by a J-point elevation of ≥2 mm, coved-type ST-segment elevation, and negative T wave in one or more right precordial leads, called type 1 Brugada ECG. Risk stratification is particularly difficult in asymptomatic cases. Patients who have experienced documented VF are generally recommended to receive an implantable cardioverter defibrillator to lower the likelihood of sudden death due to recurrent episodes. However, for asymptomatic individuals, the most appropriate course of action remains uncertain. Accurate risk prediction is critical to avoiding premature deaths and unnecessary treatments. Due to the challenges associated with experimental research on human cardiac tissue, alternative techniques such as computational modeling and deep learning-based artificial intelligence (AI) are becoming increasingly important. This study introduces a vision transformer (ViT) model that leverages 12-lead ECG images to predict potentially fatal arrhythmic events in BrS patients. This dataset includes a total of 278 ECGs, belonging to 210 patients which have been diagnosed with Brugada syndrome, and it is split into two classes: event and no event. The event class contains 94 ECGs of patients with documented ventricular tachycardia, ventricular fibrillation, or sudden cardiac death, while the no event class is composed of 184 ECGs used as the control group. At first, the ViT is trained on a balanced dataset, achieving satisfactory results (89% accuracy, 94% specificity, 84% sensitivity, and 89% F1-score). Then, the discarded no event ECGs are attached to additional 30 event ECGs, extracted by a 24 h recording of a singular individual, composing a new test set. Finally, the use of an optimized classification threshold improves the predictions on an unbalanced set of data (74% accuracy, 95% negative predictive value, and 90% sensitivity), suggesting that the ECG signal can reveal key information for the risk stratification of patients with Brugada syndrome. Full article

Purpose: CT-TAVI is a critical component of pre-TAVI assessment. The conventional method, retrospective ECG-gated scan, covering a complete cardiac cycle, measures the annulus during optimal systolic phases. Recently, prospective ECG-triggered scans acquiring images at a specific interval of the cardiac cycle were evaluated, allowing faster acquisition and lower contrast doses. Moreover, these scans might be beneficial for elderly patients, reducing the need for breath-holding and easing cooperation requirements. Still, their impact on annular measurement and procedural success has yet to be fully evaluated. Methods: This retrospective, single-center study included 419 patients who underwent CT-TAVI scans, by either prospective or retrospective scanning methods. Baseline data and calculated surgical risk scores were collected, with propensity score matching performed, followed by univariate analysis, Cox regression, and multivariable regression analysis. Results: A total of 171 patient pairs were generated via propensity score matching, ensuring that both groups had similar distributions of age (81 ± 8 years), sex (55% males), and baseline comorbidities. The patients in the prospective ECG-triggered group were exposed to a smaller amount of contrast material (40.0 ± 12 mL vs. 70.0 ± 48 mL, p < 0.001) and radiation (4.4 ± 3.6 mSv vs. 8.0 ± 10.3 mSv, p < 0.001). The prospective ECG-triggered group had a smaller aortic annulus area and diameter (426.6 ± 121.0 mm² vs. 469.1 ± 130.8 mm², p = 0.006 and 23.3 ± 3.2 mm vs. 24.5 ± 3.6 mm, p = 0.004) but no excess paravalvular leak was observed. Multivariable analysis showed no significant differences in mortality and composite endpoints between the two groups after 23 months of follow-up. Conclusion: Prospective ECG-triggered, ultra-fast, low-dose, high-pitch scan protocol, used in selected patients offers comparable safety and clinical procedural outcomes along with time and contrast savings. Full article

Background: Transforming one-dimensional (1D) biomedical signals into two-dimensional (2D) images enables the application of convolutional neural networks (CNNs) for classification tasks. In this study, we investigated the effectiveness of different 1D-to-2D transformation methods to classify electrocardiogram (ECG) and electroencephalogram (EEG) signals. Methods: We select five transformation methods: Continuous Wavelet Transform (CWT), Fast Fourier Transform (FFT), Short-Time Fourier Transform (STFT), Signal Reshaping (SR), and Recurrence Plots (RPs). We used the MIT-BIH Arrhythmia Database for ECG signals and the Epilepsy EEG Dataset from the University of Bonn for EEG signals. After converting the signals from 1D to 2D, using the aforementioned methods, we employed two types of 2D CNNs: a minimal CNN and the LeNet-5 model. Our results indicate that RPs, CWT, and STFT are the methods to achieve the highest accuracy across both CNN architectures. Results: These top-performing methods achieved accuracies of 99%, 98%, and 95%, respectively, on the minimal 2D CNN and accuracies of 99%, 99%, and 99%, respectively, on the LeNet-5 model for the ECG signals. For the EEG signals, all three methods achieved accuracies of 100% on the minimal 2D CNN and accuracies of 100%, 99%, and 99% on the LeNet-5 2D CNN model, respectively. Conclusions: This superior performance is most likely related to the methods’ capacity to capture time–frequency information and nonlinear dynamics inherent in time-dependent signals such as ECGs and EEGs. These findings underline the significance of using appropriate transformation methods, suggesting that the incorporation of time–frequency analysis and nonlinear feature extraction in the transformation process improves the effectiveness of CNN-based classification for biological data. Full article

Genetically determined left ventricular excessive trabeculation (LVET) has a wide clinical spectrum ranging from asymptomatic subjects to severe heart failure with arrhythmias and thromboembolic events. Unlike other cardiomyopathies, the relatives of LVET patients never reach the spotlight of guidelines and clinical practice, although these family members can be often affected by these conditions. Thus, we aimed to investigate the relatives of LVET by multidimensional analysis, such as genetic testing, ECG and cardiac ultrasound (ECHO). We included 55 blood relatives from the family of 18 LVET patients (male = 27, age = 44 ± 20.8y), who underwent anamnesis registration. With Sanger sequencing, the relatives were classified into genetically positive (GEN-pos) and unaffected (GEN-neg) subgroups. In addition to regular ECG parameters, Sokolow-Lyon Index (SLI) values were calculated. 2D ECHO images were analysed with TomTec Arena, evaluating LV volumetric, functional (EF) and strain parameters. Individuals were categorized into JENNI-pos and JENNI-neg morphological subgroups according to the Jenni LVET ECHO criteria. Family history showed frequent involvement (arrhythmia 61%, stroke 56%, syncope 39%, sudden cardiac death 28%, implanted device 28%), as well as personal anamnesis (subjective symptoms 75%, arrhythmias 44%). ECG and ECHO parameters were within the normal range. In terms of genetics, 78% of families and 38% of relatives carried the index mutation. LV_SLI and QT duration were lower in the GEN-pos group; ECHO parameters were comparable in the subgroups. Morphologically, 33% of the relatives met Jenni-LVET criteria were genetically affected and showed lower LV_EF values. The frequently found genetic, morphological and clinical involvement may indicate the importance of screening and, if necessary, regular follow-up of relatives in the genetically affected LVET population. Full article

Background/Objectives: Diffuse myocardial fibrosis and altered deformation are relevant prognostic factors in aortic stenosis (AS) patients. The aim of this exploratory study was to investigate the relationship between myocardial strain, and myocardial extracellular volume (ECV) in patients with severe AS with a photon-counting detector (PCD)-CT. Methods: We retrospectively included 77 patients with severe AS undergoing PCD-CT imaging for transcatheter aortic valve replacement (TAVR) planning between January 2022 and May 2024 with a protocol including a non-contrast cardiac scan, an ECG-gated helical coronary CT angiography (CCTA), and a cardiac late enhancement scan. Myocardial strain was assessed with feature tracking from CCTA and ECV was calculated from spectral cardiac late enhancement scans. Results: Patients with cardiac amyloidosis (n = 4) exhibited significantly higher median mid-myocardial ECV (48.2% versus 25.5%, p = 0.048) but no significant differences in strain values (p > 0.05). Patients with prior myocardial infarction (n = 6) had reduced median global longitudinal strain values (−9.1% versus −21.7%, p < 0.001) but no significant differences in global mid-myocardial ECV (p > 0.05). Significant correlations were identified between the global longitudinal, circumferential, and radial strains and the CT-derived left ventricular ejection fraction (EF) (all, p < 0.001). Patients with low-flow, low-gradient AS and reduced EF exhibited lower median global longitudinal strain values compared with those with high-gradient AS (−15.2% versus −25.8%, p < 0.001). In these patients, the baso-apical mid-myocardial ECV gradient correlated with GLS values (R = 0.28, p = 0.02). Conclusions: In patients undergoing PCD-CT for TAVR planning, ECV and GLS may enable us to detect patients with cardiac amyloidosis and reduced myocardial contractility Full article

Billions of paper Electrocardiograms (ECGs) are recorded annually worldwide, particularly in the Global South. Manual review of this massive dataset is time-consuming and inefficient. Accurate digital reconstruction of these records is essential for efficient cardiac disease diagnosis. This paper proposes a systematic framework for digitizing paper ECGs with 12 symmetrically distributed leads and identifying abnormal samples. This method consists of three main components. First, we introduce an adaptive rotated convolution network to detect the positions of lead waveforms. By exploiting the symmetric distribution of 12 leads, a novel loss is proposed to improve the detection model’s performance. Second, image processing techniques, including denoising and connected component analysis, are employed to digitize ECG waveforms. Finally, we propose a transformer-based classification method combined with a state space model. Our process is evaluated on a large synthetic dataset, including ECG images characterized by rotations, noise, and creases. The results demonstrate that the proposed detection method can effectively reconstruct paper ECGs, achieving an 11% improvement in SNR compared to the baseline. Moreover, our classification model exhibits slightly higher performance than other counterparts. The proposed approach offers a promising solution for the automated analysis of paper ECGs, supporting clinical decision-making. Full article

D-transposition of the great arteries (D-TGA) is a common cyanotic critical congenital heart disease. An arterial switch operation (ASO) with/without a ventricular septal defect (VSD) closure is the preferred surgical approach, with an added challenge when an intramural coronary artery (IMC) is present (1), with a reported increased incidence of postoperative complications and mortality (2,3). We present our recent D-TGA with intramural coronary artery (TGA-IMC) experience, focusing on the salient features identified on echocardiography, computed tomography (CT) angiography, and invasive angiograms, as well as variations in ASO surgical techniques for repair. Diagnostic imaging evaluation allowed for identification of the lesion, as well as planning for and undertaking of two different surgical approaches. While the two patients had differing immediate postoperative courses, both were asymptomatic at discharge, with normal biventricular systolic function. Our experience demonstrates that the suspicion for a coronary anomaly in TGA can be raised prenatally and confirmed postnatally with focused trans-thoracic echocardiography and ECG-gated CT angiogram evaluation while also aiding in operative planning. Moreover, suggesting further exploration of the optimal surgical technique for the repair of TGA-IMC. Full article

Background: To validate the automated quantification of cardiac chamber volumes and myocardial mass on non-contrast chest CT using cardiac MR (CMR) as a reference. Methods: We retrospectively included 53 consecutive patients who received non-contrast chest CT and CMR within three weeks. A deep learning model created cardiac segmentations on axial soft-tissue reconstructions from CT, covering all four cardiac chambers and the left ventricular myocardium. Segmentations on CMR cine short-axis and long-axis images served as a reference. Standard estimates of diagnostic accuracy were calculated for ventricular volumes at end-diastole and end-systole (LVEDV, LVESV, RVEDV, RVESV), left ventricular mass (LVM), and atrial volumes (LA, RA) at ventricular end-diastole. A qualitative assessment noted segmentation issues. Results: The deep learning model generated CT measurements for 52 of the 53 patients (98%). Based on CMR measurements, the average LVEDV was 166 ± 64 mL, RVEDV was 144 ± 51 mL, and LVM was 115 ± 39 g. The CT measurements correlated well with CMR measurements for LVEDV, LVESV, and LVM (ICC = 0.85, ICC = 0.84, and ICC = 0.91; all p < 0.001) and RVEDV and RVESV (ICC = 0.79 and ICC= 0.78; both p < 0.001), and moderately well with LA and RA (ICC = 0.74 and ICC = 0.61; both p < 0.001). Absolute agreements likewise favored LVEDV, LVM, and RVEDV. ECG-gating did not relevantly influence the results. The CT results correctly identified 7/15 LV and 1/1 RV as dilated (one and six false positives, respectively). Major qualitative issues were found in three cases (6%). Conclusions: Automated cardiac chamber volume and myocardial mass quantification on non-contrast chest CT produced viable measurements in this retrospective sample. Relevance Statement: An automated cardiac assessment on non-contrast chest CT provides quantitative morphological data on the heart, enabling a preliminary organ evaluation that aids in incidentally identifying at-risk patients who may benefit from a more targeted diagnostic workup. Full article