Search Results (1,604)

Accurate R-peak detection in electrocardiogram (ECG) signals is fundamental for cardiovascular analysis. However, most existing methods address differences in sampling frequency (f_s) through signal resampling or transfer learning, which may alter the temporal definition of annotated events. In this study, we propose a f_s consistent framework for ECG R-peak detection that avoids both resampling and retraining. The proposed method is based on low-sampling morphological learning combined with physiological temporal constraints (PTC). A lightweight classifier based on Extreme Gradient Boosting (XGB) was trained on 128-Hz ECG data from the MIT-BIH Normal Sinus Rhythm Database to learn local morphological structures, and feature extraction is defined in milliseconds with time-normalized derivatives to ensure consistency across f_s. The trained model is directly applied to higher-f_s datasets (360 Hz, 500 Hz, and 1000 Hz) without modification. Final peak locations are determined through deterministic processing, including PTC and local snap processing. Experimental results demonstrated that the proposed method achieved stable detection performance across multiple sampling frequencies. When evaluated in a sample-wise manner, the proposed method achieved mean F1-scores of 0.885 on MIT-BIH Arrhythmia Database (360 Hz), 0.848 on Lobachevsky University Electrocardiography Database (LUDB, 500 Hz, sinus rhythm), 0.837 on LUDB (500 Hz, arrhythmia), and 0.953 on PTB Diagnostic ECG Database (1000 Hz), without any resampling or retraining. The integration of probabilistic candidate detection and deterministic temporal alignment enables consistent peak localization under cross-frequency conditions. These findings demonstrate that augmenting machine learning with deterministic decision mechanisms provides a principled framework for f_s-consistent ECG peak detection. Full article

Background: Small animal models are essential for atrial fibrillation (AF) research. Researchers in AF use an electrocardiogram (ECG), echocardiography and invasive electrophysiology study (EPS) to assess atrial structural and electrical remodeling. In relatively smaller cardiac structures and rapid heart rates, the examination can be challenging without special tools designed for animal study. Moreover, conventional invasive EPSs often cause significant trauma, alter autonomic tone, and limit longitudinal evaluations. This study aimed to evaluate the feasibility of repurposing hospital-grade medical devices for the non-invasive, multi-modality assessment of atrial myopathy in a rat model of metabolic syndrome (MetS). Methods: A total of 12 male Sprague-Dawley rats underwent the multi-modality assessment. Structural remodeling was evaluated using hospital-grade echocardiography (8–12 MHz) to measure left atrial (LA) dimensions and volume. Surface ECG was used to determine P-wave duration. Electrical remodeling and AF inducibility were assessed using transesophageal pacing (TEP)-based EPS, evaluating the atrial effective refractory period (AERP), sinus node recovery time (SNRT), and response to rapid atrial burst pacing. Results: The protocols showed high procedural safety (survival rate 91.67%) and successfully characterized atrial myopathy. Surface ECG showed marked intra-atrial conduction delay with prolonged P-wave duration in the MetS group (30.17 ± 4.62 vs. 22.33 ± 1.86 ms, p < 0.05). Echocardiography revealed signs of structural remodeling in the MetS group, evidenced by marked prolonged Isovolumic Relaxation Time (IVRT: 35.602 ± 3.043 vs. 19.187 ± 3.631 ms; p < 0.001) and increased Left Atrial Area (0.223 ± 0.0556 vs. 0.134 ± 0.033; p = 0.007). Furthermore, TEP-based EPS quantified electrical remodeling. The MetS group had shorter AERP (73.33 ± 10.33 ms vs. 120.00 ± 34.06 ms; p = 0.010) and Corrected SNRT (100.67 ± 53.98 ms) versus controls (208.33 ± 76.97 ms; p = 0.018). The MetS group exhibited a higher absolute AF inducibility rate (50%, three out of six rats) compared to the SH group (33.3%, two out of six rats). Conclusions: The integration of surface ECG, echocardiography, and TEP-based EPS provides a safe, highly reproducible, and comprehensive method for evaluating both structural and electrical components of atrial myopathy in small animal models, allowing for robust longitudinal studies. Full article

Background/Objectives: Electrocardiogram (ECG)-based myocardial infarction (MI) classifiers may achieve high internal validation performance but show reduced performance when applied to data from another source. The task is a controlled binary healthy-versus-MI benchmark and is not intended to represent real-world chest-pain triage or autonomous clinical deployment. This study evaluated the source-only transportability of engineered 12-lead ECG feature families for binary healthy-versus-MI classification across a cardiologist-annotated hospital dataset and PTB-XL. Methods: The hospital dataset contained 1749 usable recordings from 1434 patients after excluding 206 broken-data records, with 1550 Healthy and 199 MI recordings. The matched PTB-XL binary subset contained 14,982 recordings from 13,436 patients, with 9513 Healthy and 5469 MI recordings. Eleven engineered feature families and five classifier families were compared under preprocessing, patient-aware splitting, source-validation hyperparameter and threshold selection, and bootstrap uncertainty estimation. The reported leading rows are the highest observed configurations in a prespecified benchmark grid, not locked clinical models. Results: Internal performance was higher than strict source-only transfer performance. In the hospital dataset, fiducial interval descriptors with Extra Trees reached balanced accuracy 0.775 and receiver operating characteristic area under the curve (ROC-AUC) 0.855. In PTB-XL, a broad hybrid feature bank with ST-segment information and XGBoost reached a balanced accuracy of 0.898 and ROC-AUC of 0.965. Strict source-only transfer was weaker and asymmetric: the highest observed balanced accuracy was 0.580 for hospital-to-PTB-XL transfer and 0.632 for PTB-XL-to-hospital transfer. Ranking transportability and operating-threshold transportability diverged, most notably for hospital-to-PTB-XL transfer, where ROC-AUC was 0.774 but sensitivity at the source-selected threshold was only 0.164. A secondary target-threshold analysis improved balanced accuracy to 0.682 and 0.640, respectively, but this used target labels only to re-select the operating threshold and was not a strict source-only result. Conclusions: The findings indicate a transportability gap: PTB-XL-to-hospital transfer was more balanced than hospital-to-PTB-XL transfer, but neither direction achieved performance comparable to internal validation. The source-only operating-point results are not acceptable for clinical MI screening or decision support without additional calibration, target-setting validation, and prospective assessment. Full article

Current acute coronary syndrome (ACS) care relies on sequential, single-modality diagnostics, in which the electrocardiogram, the troponin trajectory, and the coronary angiogram are interpreted independently rather than as a joint signal. This narrative review maps rather than pools the evidence. We selectively searched PubMed, EMBASE, Cochrane CENTRAL, and Web of Science (January 2015–February 2026); study selection was performed by a single reviewer, without duplicate screening, a PRISMA flow diagram, or a formal risk-of-bias assessment. The three key findings are as follows: A machine learning-enabled electrocardiogram (ECG) for diagnosing occlusion due to myocardial infarction achieved an AUC of 0.938 (95% CI = 0.924–0.951) on data not seen during training and correctly diagnosed 42% of patients that expert interpreters missed. A machine learning-enabled high-sensitivity troponin interpretation method, CoDE-ACS, reported an AUC of 0.953 and increased the number of patients ruled out at initial evaluation from 27% to 61%. Angiographically derived physiological methods produced conflicting results—quantitative flow ratios reduced major adverse cardiovascular events (MACE) in the FAVOR III China trial (HR 0.65), but in FAVOR III Europe the angiography-derived approach did not prove non-inferior to FFR; if anything, QFR guidance led to more events (6.7% vs. 4.2%, an event rate about 60% higher in the QFR arm; HR 1.63; 95% CI 1.11–2.41). There was no difference between FFR-angio and FFR in the ALL-RISE trial. These are diagnostic-accuracy and prognostic-association findings; no trial has yet shown that AI-guided ACS care reduces death, reinfarction, or ischemia-driven revascularization. Full article

Wearable electrocardiogram (ECG) monitoring enables continuous cardiovascular assessment, yet signals acquired in ambulatory environments are inevitably corrupted by baseline wander, electrode motion artifacts, and muscle interference, which obscure diagnostically critical waveform features. Existing deep learning denoisers rely on heuristic attention mechanisms and time-domain-only losses, lacking principled control over what information the network retains or discards. To address this limitation, we propose EFIB-Net, an information bottleneck-guided multi-resolution network for robust ECG denoising. The framework introduces two complementary components: an efficient frequency-guided attention module that derives temporal attention weights directly from the energy distribution of parallel multi-resolution convolutional branches, requiring only four learnable parameters while providing physically interpretable feature selection that naturally highlights QRS complexes, and a variational information bottleneck constraint at the encoder–decoder bottleneck that forces the latent representation to retain only reconstruction-relevant information and discard noise, guided by a spectral–temporal composite loss. To the best of our knowledge, we are among the first to explicitly introduce the information bottleneck principle into deep-learning-based ECG signal denoising. Experiments on the MIT-BIH Arrhythmia Database show that EFIB-Net outperforms ten traditional and deep learning baselines across four standard metrics—signal-to-noise ratio (SNR), root mean square error, percentage root-mean-square difference, and correlation coefficient; at an input SNR of −5 dB it reaches $8.12$ dB output SNR, surpassing the strongest attention-based competitor by 1.77 dB ($p<0.01$) while using only 0.45 M parameters and 10.8 ms inference latency per segment; downstream evaluation further demonstrates that the denoised signals achieve 99.18% R-peak detection sensitivity and 91.26% heartbeat classification F1-score, both within approximately one percentage point of the clean-signal upper bound, making it practical for real-time cardiac monitoring on resource-constrained wearable devices. Zero-shot cross-database evaluation on the QT Database further confirms generalizability, with only 0.54 dB degradation without retraining. Full article

Deep learning has achieved remarkable progress in electrocardiogram (ECG) analysis, but its heavy dependence on labeled data greatly increases annotation cost. This work proposes wall-aligned vector embedding (WAVE), a self-supervised learning framework that effectively extracts prior knowledge from unlabeled ECGs to reduce reliance on labels. WAVE fully leverages the diversity, synergy, and lead correlation of multi-lead ECGs by explicitly incorporating the correspondence between ECG leads and cardiac walls. Specifically, a multi-branch network captures lead-wise diversity; wall-wise synergy is modeled by concatenating leads from the same wall and projecting them via shared projection; and a dual alignment task is designed to learn correlations both within and across cardiac walls. Experimental results demonstrate that WAVE consistently surpasses all baselines under various evaluation settings, and maintains strong performance even when only a small fraction of labeled ECGs is available. Furthermore, components such as dual alignment, shared projection, wall-based concatenation, and mean target embedding are empirically verified to significantly enhance pretraining quality. In summary, WAVE learns highly informative ECG representations from unlabeled data, enabling low-cost and label-efficient ECG analysis for real-world cardiovascular diagnostics. Full article

Background/Objectives: Anthracycline-induced cardiotoxicity remains a major challenge in cancer treatment, and researchers are showing interest in artificial intelligence (AI) to improve the prediction and detection of cancer therapy-related cardiac dysfunction (CTRCD). Current surveillance strategies rely mainly on left ventricular ejection fraction and, more recently, global longitudinal strain. Methods: The present study was designed to evaluate cardiac performance in a rat model of doxorubicin-induced cardiotoxicity and empagliflozin-mediated cardioprotection using a machine learning-based analytical framework. Eighteen adult male Sprague–Dawley rats were assigned to five experimental groups. We aimed to quantify ventricular wall dynamics and contractility using an advanced image-processing and object-detection model that has not been previously used to distinguish normal from impaired cardiac kinetics. During real-time recording, simultaneous electrocardiogram monitoring was performed, enabling direct correlation between deep learning-based ventricular wall motion metrics and cardiac electrical activity. The cardioprotective effects of empagliflozin were further validated by immunofluorescence staining (cTnI, vimentin, α-SMA, and Cx43) of rat cardiomyocytes and paraffin-embedded cardiac tissue, demonstrating attenuation of cellular injury and structural remodeling. Results: The integrated analysis of cardiac kinetic patterns derived via machine learning distinguishes not only extreme cardiotoxicity, but also tracks a graded pattern consistent with ECG-derived severity and treatment-related functional preservation. These findings indicate that the algorithm captures the gradient of empagliflozin’s cardioprotective effect within this internally validated preclinical setting. Additionally, immunofluorescence results validated the benefits of SGLT2 inhibition on myocardial integrity. Conclusions: The novelty of the present work lies at the intersection of advanced cardiac kinetic analysis using AI, preclinical modeling, and SGLT2-mediated cardioprotection in cardio-oncology. Full article

Electrocardiogram (ECG) and photoplethysmogram (PPG) signals are widely used noninvasive methods for assessing cardiovascular activity and provide complementary information about the cardiac cycle. ECG records cardiac electrical activity, whereas PPG records optically detected blood-volume changes in peripheral tissue. This paper describes a synchronized ECG-PPG dataset collected from 148 apparently healthy subjects under a controlled seated protocol at rest and during post-exercise recovery after two treadmill-running conditions. Signals were acquired using a custom card-type handheld finger-contact prototype that records single-lead ECG and dual-wavelength PPG at 660 nm and 940 nm concurrently. The dataset contains 444 condition-specific records, with each subject contributing one seated resting record, one seated recovery record after light treadmill running, and one seated recovery record after moderate treadmill running. Both raw ADC-count signals and preprocessed signals are provided, and the accompanying software and example code are publicly available. The dataset is intended for research on synchronized ECG-PPG signal analysis, waveform-quality assessment, controlled post-exercise recovery physiology, and exploratory PPG-to-ECG reconstruction under controlled conditions. It should not be interpreted as a free-living wearable dataset or as clinical diagnostic ECG ground truth without external validation. Full article

Fatigue detection in remote tower air traffic controllers is important for maintaining operational safety under sustained visual monitoring and high cognitive workload. This study proposes MFD-Net, a dual-stream multimodal fusion framework using eye-tracking and electrocardiogram (ECG) signals. The model separately encodes eye-tracking and ECG-derived temporal inputs, incorporates an ECG-derived RMSSD expert feature, and performs lightweight late fusion for fatigue-state classification. Under the mixed-subject random-window protocol, MFD-Net achieved an Accuracy of 85.20%, a Recall of 83.33%, and an AUC of 0.9337. Because overlapping windows from the same participant and scenario could appear in both training and test sets, this result should be interpreted as a potentially optimistic within-distribution estimate. Under the stricter zero-shot leave-one-subject-out (LOSO) protocol, performance decreased substantially, with an Accuracy of $70.95\pm 21.59\%$, a Recall of $22.98\pm 36.30\%$, and an AUC of $0.6025\pm 0.2984$. This low zero-shot Recall indicates limited subject-independent fatigue-detection capability. Lightweight target-subject calibration and sequential probability aggregation improved adaptation and temporal stability, although the calibration results should be interpreted cautiously because random target-subject windows were used for fine-tuning. These findings suggest that eye-tracking and ECG fusion are promising under controlled conditions, while practical deployment requires deployment-oriented calibration protocols, recall-oriented optimization, and further real-world validation. Full article

The high prevalence of cardiovascular diseases and the extensive application wearable electrocardiogram (ECG) devices for long-term monitoring have posed significant challenges for the transmission, storage, and real-time processing of massive amounts of ECG data. Consequently, efficient ECG compression and reconstruction have become a research priority in remote ECG monitoring. Traditional compressed sensing is complex and has high computational overhead, while single deep learning models cannot simultaneously extract local waveforms and model temporal dependencies. To address these shortcomings in the reconstruction process, this paper presents a CNN-LSTM-Attention hybrid model. This model utilizes a convolutional neural network (CNN) to capture local ECG waveform features, employs a long short-term memory (LSTM) network to learn long-term temporal dependencies, and introduces an attention mechanism to weight and fuse key diagnostic features, enabling accurate focus on key components including the QRS complex and ST segment. Experimental results on the MIT-BIH Arrhythmia dataset demonstrate that across the full compression range of 0.1–0.9, the proposed model achieves favorable comprehensive performance. Its PRD is stabilized at 10–12%, the SNR stays above 20 dB, and the RMSE is mostly lower than 0.25 mV. In terms of reconstruction accuracy and stability, our model outperforms the single CNN and CNN-LSTM models by a large margin. Full article

Background and Objectives: During intense bouts of physical activity, the body of athletes is subjected to stress and sometimes this can lead to adverse events such as injuries or more severe organ dysfunction, like sudden cardiac death. Several markers are being studied to properly assess the level of physical stress that exercises have on the body and one of them is amylase. Materials and Methods: We evaluated 19 licensed football players using basic cardiovascular procedures, i.e., resting 12-lead electrocardiogram (ECG) and trans-thoracic echocardiography (TTE) and performing a cardiopulmonary exercise testing (CPET). Resting (T0) serum and salivary amylase levels were measured, as were immediately post-effort (T1) serum values and 10 min (T2) and 30 min post-CPET (T3) salivary values. Results: Both serum and salivary levels showed correlations with several TTE and CPET parameters. Only T2 salivary amylase levels did not show any correlations with the other parameters, while also no correlations could be established between serum and salivary determinations. Conclusions: Serum and salivary amylase determinations show potential in athlete evaluation even from a cardiovascular risk standpoint since they displayed several correlations with both TTE and CPET parameters, but as part of a more complex protocol. Salivary determinations cannot fully substitute serum measurements. Further studies on larger groups are required. Full article

Ischemic Heart Disease (IHD) is the leading cause of cardiovascular mortality worldwide. Early detection of ischemic changes using electrocardiogram (ECG) signals is vital for timely intervention and enhanced clinical outcomes. However, the diagnosis of IHD varies significantly between men and women. Women often present with atypical symptoms, and their cardiovascular risk is frequently underestimated, which leads to delayed diagnosis. Also, existing approaches face challenges in subtle early-stage abnormalities, single-lead ECG presentation, and the limited interpretability of deep learning models. These cause significant challenges to the accurate diagnosis of IHD. To address these, this study proposes a gender-aware framework, Gender-Aware Morphology Encoder Network (GAMENet), for early ischemic heart disease detection using 12-lead ECG signals with clinical metadata. A novel GAMENet is developed using the PTB-XL database. The Adaptive Morphology Deviation Encoder (AMDE) through Morphology Segment Extraction (MSEG-R) using R-Peak anchoring, isolates clinically relevant waveform components (P-wave, QRS complex, ST-segment, and T-wave) from the preprocessed ECG signals. The feature vector of morphology features is passed through dense layers with dropout regularization and a SoftMax classifier. Statistical and comparative analysis ensures that the proposed framework enables accurate IHD classification and improved interpretability. Full article

Single-lead electrocardiogram (ECG) is widely used in wearable devices for atrial fibrillation (AF) screening. Nevertheless, subtle pathological characteristics like P-waves and f-waves in practical signals are vulnerable to noise contamination. Meanwhile, the scarcity of high-quality annotated abnormal data instances leads to severe class imbalance. To mitigate these issues, we present an end-to-end framework designed for arrhythmia diagnosis using single-lead ECG signals, which integrates quality-aware data augmentation with a Peak-Enhanced attention mechanism. First, to mitigate the problem of data imbalance, a Quality-Aware Generative Adversarial Network (QA-GAN) is designed. This network integrates a signal quality evaluation module based on signal kurtosis, together with a dynamic soft-label training scheme, guiding the generator to prioritize learning high-quality morphological features, thereby synthesizing high-fidelity minority class samples. Second, to accurately capture subtle pathological features in electrocardiograms, a Peak-Enhanced Attention Convolutional Network (PEAC-Net) classification model is proposed. This model incorporates a Peak-Enhanced Attention (PE-Att) module, which employs learnable derivative convolutional kernels to precisely identify the transition points in the ECG signal. Furthermore, by integrating one-dimensional multi-scale dilated convolution (DSGC1D) with bidirectional LSTM, the model achieves effective capturing of both fine-grained local morphological features and long-range global rhythm patterns. Experimental results on the PhysioNet 2017 dataset indicate that the presented model attains an accuracy of 0.902 and a macro-F1 score of 0.880, respectively, outperforming other state-of-the-art models and also exhibiting robust data adaptability on the MIT-BIH dataset. Full article

Background and Objectives: Premature ventricular contractions are among the most common arrhythmias encountered in clinical practice. However, this disorder can be associated with arrhythmia-induced cardiomyopathy or be the first sign of primary myocardial diseases. Certain morphologies of premature ventricular contractions are associated with a higher risk for sudden arrhythmia and cardiac dysfunction in the adult population. There is data on the clinical value and significance of the contraction morphology in adults, but there is a lack of such data for children. Materials and Methods: This observational prospective study of pediatric outpatients with premature ventricular contractions was conducted at Vilnius University Hospital Santaros Clinics. Inclusion criteria comprised children aged 3–17 years with more than 5% premature ventricular contractions over 24 h. Exclusion criteria included previously diagnosed congenital heart defects and cardiomyopathies, channelopathies, or the presence of any acute condition. The electrocardiographic morphology and measurements were assessed, analyzed, and described in this study. Results: The electrocardiograms of 80 patients were analyzed according to the ECG-estimated morphology of the arrhythmia complex, arrhythmic QRS complex duration, ratio with the normal QRS complex, and maximum deflection index in V5–V6 derivations. Cardiac MRI abnormalities (8 of 30 MRI studies) was reliably associated with a PVC duration of >150 ms and the maximal amount of extrasystoles per 24 h, with a median amount of 29.6%. A long postcoupling interval (>0.9 s) was associated with PVC progression. Conclusions: In this exploratory pediatric cohort, wider PVC QRS duration and higher maximal PVC burden were associated with ventricular MRI abnormalities, while longer postcoupling interval was associated with PVC progression. Full article

Cardiovascular diseases impose a substantial global health burden and often require timely detection, creating strong demand for real-time electrocardiogram (ECG) monitoring on resource-constrained devices. Although portable single-lead wearable ECG devices are valuable for daily monitoring, their diagnostic performance is limited by spatial information loss and hardware constraints. Moreover, conventional lightweight models lack interpretable analysis beyond coarse classification. This study proposes an edge–cloud collaborative ECG-assisted analysis method combining lightweight ECG model distillation with large language models. At the algorithmic level, a cross-lead distillation framework transfers knowledge from a 12-lead InceptionTime–Transformer teacher to an ultra-lightweight single-lead student via a hybrid loss integrating hard-label, temperature-scaled soft-label, and auxiliary multi-label objectives. At the system level, a three-layer architecture integrates edge-side real-time screening with cloud-side report generation through a LoRA-fine-tuned Qwen3-8B model. Experiments on PTB-XL show that, under 123.7× parameter compression and 12-to-1 lead reduction, the student retains 92.8% of the teacher’s Macro-F1 and 94.7% of its AUC-ROC. After 8-bit integer (INT8) quantization, the TFLite file is 20.8 KB; QEMU-based Cortex-M4 simulation shows approximately 63.0 KB SRAM usage and 11.6 ms latency, suggesting potential on-device deployment under simulated conditions. Validation on physical hardware—including power consumption, BLE latency, and motion artifacts—remains necessary. Full article