Irregular Rhythm Research Articles

THE PREDICTION OF HEART DISEASE USING MACHINE LEARNING ALGORITHMS

Heart disease threatens the lives of around one individual per minute, establishing it as the foremost cause of mortality in the contemporary era. A wide range of individuals over the globe has encountered the intricacies associated with cardiovascular illness. Various factors, such as hypertension, elevated levels of cholesterol, and an irregular pulse rhythm hinder the early identification of a cardiovascular disease. In cardiology, similar to other branches of Medicine, timely and precise identification of cardiac diseases is of utmost importance. Anticipating the onset of heart failure at the appropriate moment can provide challenges, particularly for cardiologists and surgeons. Fortunately, categorisation and forecasting models can assist the medical business and provide real applications for medical data. Regarding this, Machine Learning (ML) algorithms and techniques have benefited from the automated analysis of several medical datasets and complex data to aid the medical community in diagnosing heart-related diseases. Predicting if the patient has early-stage cardiac disease is the primary goal of this paper. A prior study that worked on the Erbil Heart Disease dataset has proved that Naïve Bayes (NB) got an accuracy of 65%, which is the worst classifier, while Decision Tree (DT) obtained the highest accuracy of 98%. In this article, a comparison study has been applied using the same dataset (i.e., Erbil Heart Disease dataset) between multiple ML algorithms, for instance, LR (Logistic Regression), KNN (K-Nearest Neighbours), SVM (Support Vector Machine), DT (Decision Tree), MLP (Multi-Layer Perceptron), NB (Naïve Bayes) and RF (Random Forest). Surprisingly, we obtained an accuracy of 98% after applying LR, MLP, and RF, which was the best outcome. Furthermore, the accuracy obtained by the NB classifier differed incredibly from the one received in the prior work.

Cyanotic heart disease due to a hypoplastic right ventricle with pulmonary valve atresia, an intact ventricular septum, with atrial level shunting in a day‐old Arabian foal

AbstractAn 8‐h‐old Arabian foal was presented for evaluation of progressive weakness since birth. The foal presented laterally recumbent with pale and muddy mucous membranes, a grade III/VI left basilar murmur, and an irregularly, irregular rhythm. Transthoracic echocardiogram along with an electrocardiogram (EKG) was performed. Atrial fibrillation was recorded throughout the examination. The left and right atria were both significantly enlarged. A right to left shunt across the atrial septum was confirmed via an intravenous injection of agitated saline. The tricuspid valve was rudimentary without apparent atrioventricular flow and no ventricular septal defect was appreciated. The animal was euthanased, and necropsy found dysplasia of the tricuspid valve, a hypoplastic right ventricle with pulmonary valve atresia, and an intact ventricular septum. This condition is similar, both haemodynamically and in terms of prognosis, to tricuspid atresia, which has been described previously in Arabian foals, with the exception that there is the complete absence of connection between the atrium and ventricle in the case of tricuspid atresia. Of 14 previously reported cases of tricuspid atresia in neonatal foals, nine were Arabian foals. Further investigation into the possibility of hypoplastic right ventricle as a familial condition in the Arabian breed should be considered.

Quantitative 3D electron microscopy characterization of mitochondrial structure, mitophagy, and organelle interactions in murine atrial fibrillation

Atrial fibrillation (AF) is the most common clinical arrhythmia, however there is limited understanding of its pathophysiology including the cellular and ultrastructural changes rendered by the irregular rhythm, which limits pharmacological therapy development. Prior work has demonstrated the importance of reactive oxygen species (ROS) and mitochondrial dysfunction in the development of AF. Mitochondrial structure, interactions with other organelles such as sarcoplasmic reticulum (SR) and T-tubules (TT), and degradation of dysfunctional mitochondria via mitophagy are important processes to understand ultrastructural changes due to AF. However, most analysis of mitochondrial structure and interactome in AF has been limited to two-dimensional (2D) modalities such as transmission electron microscopy (EM), which does not fully visualize the morphological evolution of the mitochondria during mitophagy. Herein, we utilize focused ion beam-scanning electron microscopy (FIB-SEM) and perform reconstruction of three-dimensional (3D) EM from murine left atrial samples and measure the interactions of mitochondria with SR and TT. We developed a novel 3D quantitative analysis of FIB-SEM in a murine model of AF to quantify mitophagy stage, mitophagosome size in cardiomyocytes, and mitochondrial structural remodeling when compared with control mice. We show that in our murine model of spontaneous and continuous AF due to persistent late sodium current, left atrial cardiomyocytes have heterogenous mitochondria, with a significant number which are enlarged with increased elongation and structural complexity. Mitophagosomes in AF cardiomyocytes are located at Z-lines where they neighbor large, elongated mitochondria. Mitochondria in AF cardiomyocytes show increased organelle interaction, with 5X greater contact area with SR and are 4X as likely to interact with TT when compared to control. We show that mitophagy in AF cardiomyocytes involves 2.5X larger mitophagosomes that carry increased organelle contents. In conclusion, when oxidative stress overcomes compensatory mechanisms, mitophagy in AF faces a challenge of degrading bulky complex mitochondria, which may result in increased SR and TT contacts, perhaps allowing for mitochondrial Ca2+ maintenance and antioxidant production.

The association between irregularity in sleep-wake rhythm and CPAP adherence

This study aims to evaluate the association between sleep-wake rhythm regularity and continuous positive airway pressure (CPAP) adherence. We retrospectively analyzed sleep-wake rhythms with activity monitoring and CPAP adherence among obstructive sleep apnea (OSA) patients newly diagnosed and introduced to CPAP therapy at the Sapporo Hanazono Hospital from January 2018 to June 2022. Among a total of 45 patients, 10 withdrew from CPAP therapy within a year. Nineteen were classified into the good-adherence and 16 into the poor-adherence group. No significant differences were detected among the groups in apnea-hypopnea index (AHI), sleep efficiency, or subjective sleep quality, but a difference was observed in sleep latency, with the CPAP withdrawal group showing higher variability in sleep onset and lower regularity and/or amplitude in circadian behavior activity rhythm than the good-adherence group. Our results suggest that irregularities, particularly in sleep onset, and damped sleep-wake rhythm can be risk factors for CPAP withdrawal.

Tachycardia with cycle length alternans in Ebstein's anomaly.

The tachycardia which presents with regularly irregular rhythm consists of a broad set of differential diagnoses. We present a case of cycle length alternans tachycardia in a patient, with Ebstein's anomaly and describe how a diagnosis was arrived at after careful analysis of electrocardiogram and EGMs.

Comparative Profiles of the WATCHMAN™ and Amplatzer™ Cardiac Plug/Amplatzer™ Amulet™ Devices for Left Atrial Appendage Closure in Non-valvular Atrial Fibrillation: A Comprehensive Systematic Review and Meta-analysis.

Atrial fibrillation (AF) is a prevalent cardiac arrhythmia marked by irregular and frequent tachycardic rhythms in the atria, affecting 1%-2% of the general population. The WATCHMAN™ device from Boston Scientific (Marlborough, MA, USA) and the Amplatzer™ Amulet™ device from Abbott (Chicago, IL, USA) are two devices used globally for left atrial appendage closure (LAAC) in non-valvular AF. A systematic search was conducted in PubMed, the Cochrane Library, and Elsevier's ScienceDirect literature databases to identify studies comparing the WATCHMAN™ procedure with Amulet™ device implantation for LAAC in patients with AF. The analyses were conducted using the random-effects model. A total of 20 studies were identified, with 18 falling into the category of observational studies and 2 being randomized controlled trials. A total of 6310 participants were included in this meta-analysis, with 3198 individuals (50.68%) assigned to the WATCHMAN™ procedure group and 3112 individuals (49.32%) allocated to the Amplatzer™ Cardiac Plug (ACP) group. The analysis revealed a higher risk of stroke associated with the WATCHMAN™ technique (relative risk [RR], 1.14), albeit without statistical significance. Conversely, the WATCHMAN™ approach led to a significantly lower risk of cardiac death (RR, 0.44; P = .04). Notably, the risks of all-cause mortality (RR, 0.89; 95% confidence interval [CI], 0.73-1.08; I 2 = 0%; P = .25) and major bleeding (RR, 0.93; 95% CI, 0.65-1.33; I 2 = 31%; P = .70) were clinically reduced with the WATCHMAN™ procedure, although statistical significance was not achieved. Compared to Amulet™ device implantation, WATCHMAN™ device implantation decreased the risk of cardiac mortality, while the risks of stroke, systemic embolism, all-cause mortality, and major bleeding were not statistically significant.

Arrhythmia Classification from ECG Signals Using Transformers and Data Balancing Techniques

While many arrhythmias pose minimal threat, certain heart rhythm irregularities elevate the potential for stroke or heart failure. The complexity arises particularly with the supraventricular premature heartbeat which has a resemblance to a normal beat and occurs infrequently. Consequently, this research proposes a data balancing and classification technique that enhances the accuracy of identifying mentioned hard-to-classify heartbeats while maintaining robust metrics for other classes. The study introduces a deep learning framework combined with a multi-head attention transformer, for balancing – under-sampling and synthetic minority oversampling are used. To evaluate the proposed model, various experiments based on real data were conducted. The results were compared with an existing model used in chest belt heartbeat monitoring, and the results show that the transformer model achieved better performance for supraventricular premature heartbeats, at the same time reaching high overall and per-class metrics.

Rhythm-adaptive statistical estimation methods of probabilistic characteristics of cyclic random processes

The paper is devoted to the rhythm-adaptive statistical estimation methods of cyclic random processes probabilistic characteristics. The main goal of the article is development and research of rhythm-adaptive methods of cyclic random processes statistical processing. Based on the fundamental fact of invariance of probabilistic characteristics of a cyclic random process to time shifts determined by its rhythm function, rhythm-adaptive statistical estimations of initial, central and mixed moment functions of cyclic random processes of discrete parameter were obtained. The article analytically proves the unbiasedness and consistency of rhythm-adaptive statistical estimates of the mathematical expectation and the initial moment functions of the higher order of cyclic random process. By conducting a series of computer simulation experiments, it has been demonstrated, that the absolute errors of rhythm-adaptive methods of statistical estimation are significantly smaller than the corresponding errors for classical non-rhythm adaptive methods of statistical estimation of cyclic stochastic signals probabilistic characteristics. The effectiveness of using rhythm-adaptive methods of statistical estimation of probabilistic characteristics of ECG signals has been demonstrated. The results obtained in the work clearly demonstrate the essential advantages of rhythm-adaptive methods of statistical processing of cyclic stochastic signals with a variable (irregular) rhythm in comparison with non-rhythm-adaptive analogues.

Cardiac Manifestations of Wegener's Granulomatosis: A Case Report and Literature Review

Wegener's Granulomatosis is a rare systemic vasculitis. Lesions commonly occur in the upper airways, lungs, kidneys, and can also affect the skin, eyes, joints, peripheral nervous system, and the heart. Cardiac involvement is rare and can affect the endocardium, myocardium, or pericardium. Pulmonary arterial hypertension (PAH) has been rarely reported. We report the case of D.R., a 17-year-old without significant personal medical history, referred for the evaluation of new-onset exertional dyspnea. The clinical examination was normal except for an irregular heart rhythm with a pronounced pulmonary B2 sound. The ECG revealed atrial fibrillation (AF) with incomplete right bundle branch block (iRBBB). A chest X-ray showed cardiomegaly primarily affecting the atria without signs of pulmonary congestion. Precapillary PAH was suggested by echocardiography and confirmed through right heart catheterization. Routine blood tests were normal. Contrast-enhanced chest CT showed no evidence of pulmonary embolism or signs of pulmonary parenchymal involvement. Given the patient's young age, a specific vasculitis work-up was conducted, leading to a diagnosis of Wegener's Granulomatosis. Through this case report, we will present the cardiac and extracardiac manifestations of this vasculitis, as well as therapeutic management strategies.

ECG of the month

In this issue's ECG of the month, Charles Bloe presents the case of a 19-year-old man with a family history of cardiomyopathy who noticed an irregular heart rhythm on his smartwatch.

Fetal premature atrial contractions: natural course, risk factors and adverse outcome.

Fetal premature atrial contractions (PAC) are usually benign but may be associated with congenital heart defect (CHD) and tachyarrhythmia, which in turn carry a risk of cardiac failure and fetal death. We aimed to explore the frequency of adverse outcome and to identify risk factors for tachyarrhythmia in pregnancies complicated by fetal PAC. This was a retrospective cohort study of fetuses diagnosed with PAC at two academic centers in Amsterdam, The Netherlands, between 2007 and 2022. Cases with a congenital anomaly and those with a prior diagnosis of CHD or other arrhythmias were excluded. M-mode and Doppler tracings were reanalyzed and the frequency of PAC was recorded. The incidence of the following adverse outcomes was examined: underlying CHD not identified at the 20-week fetal anomaly scan, tachyarrhythmia, other arrhythmia, administration of antiarrhythmic therapy and death. Risk factors for tachyarrhythmia were analyzed using odds ratios (OR). In 24.1% of 1439 referred cases, PAC resolved before confirmation at the fetal medicine unit (FMU). Of the 939 included cases with confirmed PAC, the total incidence of adverse outcome was 6.8% (64/939). CHD was diagnosed in 14 (1.5% (95% CI, 0.9-2.5%)) cases, of which eight were diagnosed prenatally and six postnatally. Compared with baseline, the incidence of CHD was higher in the presence of fetal PAC (OR, 1.8 (95% CI, 1.0-3.3); P = 0.034). Tachyarrhythmia occurred prenatally and/or postnatally in 32 (3.4%) cases, of which eight (25.0%) showed signs of cardiac failure, and in 23 (71.9%) cases, antiarrhythmic therapy was required. No cases of tachyarrhythmia led to fetal or infant death. Risk factors for tachyarrhythmia were PAC with short runs of supraventricular tachycardia (OR, 98.7), blocked PAC (OR, 30.3), PAC in bigeminy (OR, 21.8), frequent PAC (one per 5-10 beats) (OR, 6.9), signs of cardiac failure (OR, 14.2) and the presence of a foramen ovale aneurysm (OR, 5.0). PAC are generally benign and often resolve spontaneously. However, fetuses with an irregular heart rhythm should be referred for advanced ultrasonography, which should focus on classifying the type of PAC and risk of adverse outcome. When risk factors for tachyarrhythmia are identified, weekly fetal heart-rate monitoring is advised until resolution of the PAC. In the absence of risk factors, standard obstetric care may be sufficient, with additional instructions to report reduced fetal movements. Should tachyarrhythmia or cardiac failure develop, referral back to the FMU is indicated. © 2023 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

Circulating neuropeptide Y as a biomarker in postoperative atrial fibrillation cases administered off-pump coronary bypass Graft surgery

Background and aimsPostoperative atrial fibrillation (POAF) is considered the most prevalent irregular heart rhythm after heart surgery. The cardiac autonomic nervous system significantly affects POAF, and neuropeptide Y (NPY), an abundant neuropeptide in the cardiovascular system, is involved in this autonomic regulation. The current work aimed to examine the potential association of NPY with POAF in individuals administered isolated off-pump coronary artery bypass grafting. MethodsFrom January 1 to May 31, 2020, we examined consecutive cases administered successful isolated off-pump coronary artery bypass grafting with no previously diagnosed atrial fibrillation (AF). Clinical characteristics and plasma samples were collected before surgery. NPY was quantified by enzyme-linked immunosorbent assay (ELISA) in peripheral blood, and POAF cases were identified through a 7-day Holter monitoring. ResultsAmong 120 cases with no previously diagnosed AF, 33 (27.5 %) developed POAF during hospitalization. Median NPY levels were markedly elevated in the POAF group in comparison with the sinus rhythm group (31.72 vs. 27.95, P = 0.014). Multivariable logistic regression analysis revealed age (OR = 1.135, 95%CI 1.054–1.223; P = 0.001), left atrial size (OR = 1.136, 95%CI 1.004–1.285; P = 0.043), and NPY levels in peripheral blood (OR = 1.055, 95%CI 1.002–1.111; p = 0.041) independently predicted POAF. Additionally, NPY levels were positively correlated with high-frequency (HF) (r = 0.2774, P = 0.0022) and low-frequency (LF) (r = 0.2095, P = 0.0217) components of heart rate variability. ConclusionIn summary, this study demonstrates an association between elevated NPY levels in peripheral blood before surgery and POAF occurrence.

Sleep-wake behaviors associated with cognitive performance in middle-aged participants of the Hispanic Community Health Study/Study of Latinos

ObjectivesMany sleep-wake behaviors have been associated with cognition. We examined a panel of sleep-wake/activity characteristics to determine which are most robustly related to having low cognitive performance in midlife. Secondarily, we evaluate the predictive utility of sleep-wake measures to screen for low cognitive performance. MethodsThe outcome was low cognitive performance defined as being >1 standard deviation below average age/sex/education internally normalized composite cognitive performance levels assessed in the Hispanic Community Health Study/Study of Latinos. Analyses included 1006 individuals who had sufficient sleep-wake measurements about 2years later (mean age=54.9, standard deviation= 5.1; 68.82% female). We evaluated associations of 31 sleep-wake variables with low cognitive performance using separate logistic regressions. ResultsIn individual models, the strongest sleep-wake correlates of low cognitive performance were measures of weaker and unstable 24-hour rhythms; greater 24-hour fragmentation; longer time-in-bed; and lower rhythm amplitude. One standard deviation worse on these sleep-wake factors was associated with ∼20%-30% greater odds of having low cognitive performance. In an internally cross-validated prediction model, the independent correlates of low cognitive performance were: lower Sleep Regularity Index scores; lower pseudo-F statistics (modellability of 24-hour rhythms); lower activity rhythm amplitude; and greater time in bed. Area under the curve was low/moderate (64%) indicating poor predictive utility. ConclusionThe strongest sleep-wake behavioral correlates of low cognitive performance were measures of longer time-in-bed and irregular/weak rhythms. These sleep-wake assessments were not useful to identify previous low cognitive performance. Given their potential modifiability, experimental trials could test if targeting midlife time-in-bed and/or irregular rhythms influences cognition.

Testing Textual and Territorial Boundaries in Bulat Okudzhava’s Song “And We to the Doorman: ‘Open the Doors!’”

This paper contextualizes Okudzhava’s song “And We to the Doorman” (AWD), initially marginal in the Soviet poetic mainstream. It explores its shifts in tone, irregular rhythms, colloquial language, and semi-criminal undertones. AWD’s structure, with uneven stanzas and no clear refrain, reveals underlying symmetry and recurring themes. The meter is predominantly iambic but varies. Unconventional verse endings and various rhyme schemes, including distant chains, characterize its prosody. The narrative touches on social cohesion and class conflict. The style reflects a challenging attitude toward privilege, employing rhetorical devices and indirect threats. The melody aligns with thematic elements, featuring repetitive patterns and a spoken quality. Semantically, AWD presents an ambiguous message on class struggle and moral issues. In sum, this analysis uncovers Okudzhava’s song’s formal complexities, thematic nuances, and stylistic innovations.

Acquisition and Analysis of Cardio Graphic Signal Using Various Transforms

In the modern days, cardiac problems among various age group people have increased by a large ratio. Electro cardiograph is widely used to monitor the electrical activities of the heart. With the help of ECG, we can able to easily detect and classify the irregular activities of the heart. Hence in this study, we have acquired ECG signals from ten different subjects of various age groups using the BIOPAC system. All the acquired ECG signals are pre-processed using the infinite impulse response bandpass filter. Then the most vital R wave is detected by using the peak detection method present in-built in the BIOPAC system. This R wave detection plays a very vital role in diagnosing heart rhythm irregularities, calculating the heart rate variability, recognizing the heart rate variability, and measuring the heart rate. The pre-processed ECG signal is processed using highly advanced windowing techniques to extract the power spectrum and frequency-based values of the ECG signal. The results of the Fourier transform and spectral density of the processed ECG signals are displayed in this study. The features extracted from ECG signals acquired from ten different subjects have been tabulated in this study.

Visual assessment of Korotkoff sounds improves the accuracy of a validated professional automated auscultatory blood pressure monitor KOROT P3 Accurate.

A novel automated auscultatory upper-arm cuff blood pressure (BP) monitor for office use (KOROT P3 Accurate, previously InBody BPBIO480KV), which displays Korotkoff sound curves for each BP reading was recently developed. This study investigated whether the review of Korotkoff sound curves by healthcare professionals further improves the accuracy of the device by identifying unreliable BP readings. Three observers assessed independently the morphology of Korotkoff sound curves of BP measurements obtained during an ISO 81060-2:2018 validation study, and classified them as of good, fair, or poor quality (low amplitude or sound intensity, aberrant morphology, background noise, signal artifact, auscultatory gap, irregular rhythm). The observers were blinded to the study BP measurements. Korotkoff sound curves of 255 BP readings obtained in 85 individuals were analyzed (mean age 57.3 ± 15.0 years, 53 men). Of the SBP readings 80.4/12.2/7.4% were classified as good/fair/poor, and DBP 76.9/12.2/10.9%. Inter-observer agreement in detecting poor-quality curves was 84.7/83.1% (systolic/diastolic). Of poor-quality curves, 10.5/60.7% (systolic/diastolic) clustered in the same individuals. The validation criterion 1 [mean test-reference BP difference ≤5 ± 8 (SD) mmHg] was satisfied for readings with good (0.1 ± 4.9/0.3 ± 3.8 mmHg, systolic/diastolic) and fair-quality curves (-0.4 ± 6.4/0.2 ± 5.0), but not for poor-quality ones (2.7 ± 8.8/3.6 ± 8.1). By excluding poor-quality readings (40 of 255), criterion 1 of the validation study was improved (0.2 ± 4.9/0.2 ± 3.9 versus 0.3 ± 5.5/0.6 ± 4.7 mmHg). The visual assessment of Korotkoff sounds generated during automated auscultatory BP measurement by the KOROT P3 Accurate professional monitor identifies unreliable readings and further improves the device accuracy.

Ventricular bigeminy associated with myocardial ischemia in a dog with a colonic torsion: a case report

BackgroundVentricular bigeminy due to myocardial ischemia has been reported in humans as well as in canine patients with obstructive gastrointestinal diseases. This is the first case report of ventricular bigeminy in a dog with a colonic torsion that resolved after fluid resuscitation and restoration of myocardial perfusion.Case presentationAn 11-year-old, male neutered mixed breed dog presented with a one day history of vomiting, tenesmus, and lethargy. Physical examination identified an irregular heart rhythm and intermittent pulse deficits. A ventricular arrhythmia represented by ventricular premature complexes (VPCs) organized in bigeminy, was appreciated on a 3-lead electrocardiogram (ECG) with a single lead (II) view. Abdominal radiographs confirmed a colonic torsion. Prior to anesthetic induction, ventricular bigeminy was non responsive to fentanyl or lidocaine. The patient was anesthetized and intravascular volume deficit was identified by dampened plethysmographic wave amplitude (plethysomographic variability), audible softening of the Doppler sound, and more pronounced pulse deficits. Fluid resuscitation was achieved with a combination of intravenous crystalloid and colloid fluid therapy comprising 7.2% hypertonic saline and 6% hetastarch. The patient’s cardiac rhythm converted to normal sinus after fluid resuscitation. The colonic torsion was surgically corrected. The patient recovered well from anesthesia and was ultimately discharged from the hospital 5 days later.ConclusionsThe present case report highlights that myocardial ischemia can lead to ventricular arrythmias, such as ventricular bigeminy. This is the first documented case of ventricular bigeminy in the canine patient with a colonic torsion. Assessment of patient volume status and appropriate fluid resuscitation along with continuous electrocardiogram (ECG) monitoring are vital to patient stability under general anesthesia.

Advanced Neural Network Architecture for Enhanced Multi-Lead ECG Arrhythmia Detection through Optimized Feature Extraction

Cardiovascular diseases are a pervasive global health concern, contributing significantly to morbidity and mortality rates worldwide. Among these conditions, arrhythmia, characterized by irregular heart rhythms, presents formidable diagnostic challenges. This study introduces an innovative approach utilizing deep learning techniques, specifically Convolutional Neural Networks (CNNs), to address the complexities of arrhythmia classification. Leveraging multi-lead Electrocardiogram (ECG) data, our CNN model, comprising six layers with a residual block, demonstrates promising outcomes in identifying five distinct heartbeat types: Left Bundle Branch Block (LBBB), Right Bundle Branch Block (RBBB), Atrial Premature Contraction (APC), Premature Ventricular Contraction (PVC), and Normal Beat. Through rigorous experimentation, we highlight the transformative potential of our methodology in enhancing diagnostic accuracy for cardiovascular arrhythmias. Arrhythmia diagnosis remains a critical challenge in cardiovascular care, often relying on manual interpretation of ECG signals, which can be time-consuming and prone to subjectivity. To address these limitations, we propose a novel approach that leverages deep learning algorithms to automate arrhythmia classification. By employing advanced CNN architectures and multi-lead ECG data, our methodology offers a robust solution for precise and efficient arrhythmia detection. Through comprehensive evaluation, we demonstrate the effectiveness of our approach in facilitating more accurate clinical decision-making, thereby improving patient outcomes in managing cardiovascular arrhythmias.

Multimodal ECG heartbeat classification method based on a convolutional neural network embedded with FCA

Arrhythmias are irregular heartbeat rhythms caused by various conditions. Automated ECG signal classification aids in diagnosing and predicting arrhythmias. Current studies mostly focus on 1D ECG signals, overlooking the fusion of multiple ECG modalities for enhanced analysis. We converted ECG signals into modal images using RP, GAF, and MTF, inputting them into our classification model. To optimize detail retention, we introduced a CNN-based model with FCA for multimodal ECG tasks. Achieving 99.6% accuracy on the MIT-BIH arrhythmia database for five arrhythmias, our method outperforms prior models. Experimental results confirm its reliability for ECG classification tasks.

Automated identification of atrial fibrillation from single-lead ECGs using multi-branching ResNet.

Introduction: Atrial fibrillation (AF) is the most common cardiac arrhythmia, which is clinically identified with irregular and rapid heartbeat rhythm. AF puts a patient at risk of forming blood clots, which can eventually lead to heart failure, stroke, or even sudden death. Electrocardiography (ECG), which involves acquiring bioelectrical signals from the body surface to reflect heart activity, is a standard procedure for detecting AF. However, the occurrence of AF is often intermittent, costing a significant amount of time and effort from medical doctors to identify AF episodes. Moreover, human error is inevitable, as even experienced medical professionals can overlook or misinterpret subtle signs of AF. As such, it is of critical importance to develop an advanced analytical model that can automatically interpret ECG signals and provide decision support for AF diagnostics. Methods: In this paper, we propose an innovative deep-learning method for automated AF identification using single-lead ECGs. We first extract time-frequency features from ECG signals using continuous wavelet transform (CWT). Second, the convolutional neural networks enhanced with residual learning (ReNet) are employed as the functional approximator to interpret the time-frequency features extracted by CWT. Third, we propose to incorporate a multi-branching structure into the ResNet to address the issue of class imbalance, where normal ECGs significantly outnumber instances of AF in ECG datasets. Results and Discussion: We evaluate the proposed Multi-branching Resnet with CWT (CWT-MB-Resnet) with two ECG datasets, i.e., PhysioNet/CinC challenge 2017 and ECGs obtained from the University of Oklahoma Health Sciences Center (OUHSC). The proposed CWT-MB-Resnet demonstrates robust prediction performance, achieving an F1 score of 0.8865 for the PhysioNet dataset and 0.7369 for the OUHSC dataset. The experimental results signify the model's superior capability in balancing precision and recall, which is a desired attribute for ensuring reliable medical diagnoses.