Higher order statistics for automated classification of ECG beats

Cardiac arrhythmia diagnosis method using linear discriminant analysis on ECG signals

Accurate detection of cardiac arrhythmias is crucial for preventing premature deaths. The current study employs a dual-stage Discrete Wavelet Transform (DWT) and a median filter to eliminate noise from ECG signals. Subsequently, ECG signals are segmented, and QRS regions are extracted for further preprocessing. The study considers five cardiac arrhythmias: normal beats, Premature Ventricular Contractions (PVC), Premature Atrial Contractions (PAC), Right Bundle Branch Block (R-BBB), and Left Bundle Branch Block (L-BBB) for classification. Nine distinct temporal features are extracted from the segmented QRS complex. These features are then applied to six different classifiers for arrhythmia classification. The classifiers' performance is evaluated using the MIT-BIH Arrhythmia Database (MIT-BIH AD). Support Vector Machine (SVM) and Ensemble Tree classifiers demonstrate superior performance in classifying the five different classes. Particularly, the Support Vector Machine classifier achieves high sensitivity (97.44%), specificity (99.36%), positive predictive value (97.44%), and accuracy (98.97%) with a Gaussian kernel. This comprehensive approach, integrating preprocessing, and feature extraction, holds promise for improving automatic cardiac arrhythmia classification in clinical trials.

Cardiac Arrhythmia is an abnormal heart rhythm that develops when the electrical impulses control the heart’s contraction which does not function properly. The heart can beat too fast (tachycardia), too slow (bradycardia), or in an irregular pattern. Observing ECG signal peaks and channels freehand is difficult due to their ingenious modification. Automated detection of cardiovascular abnormalities is preferred for the early diagnosis of cardiac disorders. This paper used machine learning approaches for detecting ECG abnormality utilizing a Support Vector Machine (SVM) and Cost-Sensitive Decision-Tree (CS-DT) classifier. The Empirical Mode Decomposition approach was utilized to examine the properties of R-peaks and QRS complexes in ECG signs. Various morphological characteristics are analyzed from the signal penetrated by the classifier to diagnose the irregular beats. A set of twenty-two clinically feasible features comprising temporal, morphological, and statistical were extracted from the processed ECG signals and applied to the classifier to categorize cardiovascular irregularities like Normal (N), Left Bundle Branch Block (LBBB), Right Bundle Branch Block (RBBB), Atrial Premature Beats (APB), and Premature Ventricular Contraction (PVC). The Beth Israel Hospital at Massachusetts Institute of Technology (MIT-BIH) dataset has been used for this work, where feature datasets are split into training and evaluation subsets. The training set is used to train machine learning models on the extracted features, while the evaluation set is used to assess the performance of the trained models. The evaluation metrics such as Accuracy (Acc), Sensitivity (Se), Specificity (Sp), and Positive Predictivity (Pp), are frequently used to evaluate the model’s performance in Arrhythmia detection along with classification. The simulation has been conducted using SVM and CS-DT classifier with performance for all individual class labels at a Confidence Factor (CF) of 0.5. The performance of the time and frequency domain features is merged resulting in higher classification of Sensitivity, Specificity, Positive Predictivity, and Accuracy of 89.5%, 98.11%, 87.76%, and 96.8% in SVM, 97.71%, 99.58%, 97.66%, 99.32% in CS-DT classifier in identifying the irregular heartbeats.

The long-term outcome of athletes with frequent ventricular premature complexes (VPCs) and apparently normal heart has not been fully clarified. To evaluate the clinical and prognostic significance of VPCs and the influence of continuing sports activity during follow-up, we studied 120 healthy athletes (96 men; median age 16 years) in whom frequent VPCs (>100 VPCs/24 hours) were discovered by chance during preparticipation screening. All athletes were followed up for a median of 84 months. During follow-up, 96 underwent serial 24-hour Holter recording and 62 underwent serial echocardiography. The median number of VPCs/24 hours on basal Holter was 3,760. During follow-up, 81 athletes continued sports activity, whereas 39 did not. No athlete died or developed overt heart disease. The median number of VPCs/24 hours decreased in both athletes who continued sports activity and those who did not (from 3,805 to 1,124, p <0.0001 and from 5,787 to 1,298, p <0.0001, respectively). During follow-up, left ventricular ejection fraction slightly decreased to <55% in 9 of 62 athletes who, in respect to the remaining 53, had more VPCs/24 hours both in the basal state (12,000 vs 3,880) and during follow-up (10,702 vs 1,368), and a longer follow-up (95 vs 36 months). In conclusion, (1) frequent VPCs in athletes without heart disease have a long-term benign prognostic significance, (2) sporting activity does not modify this benign outcome, (3) during follow-up, the burden of VPCs decreases whether or not subjects continue sports activity, and (4) in 14.5% of athletes, ejection fraction slightly decreases over time.

The measurement of the electrical activity of the heart can be done with electrocardiogram (ECG). Automatic arrhythmia-diagnosis systems which results in high accuracy rates for inside and outside patient are still an important area of research. The accuracy of such system depends on accuracy of the classification system. All this classification system required qualitative features for classification. This paper proposed a unique method of profiling of statistical features for selection of qualitative features through ECG waveform. The proposed approach for selection of qualitative features can classify and differentiate abnormal heartbeats and normal (NORM).Left bundle branch block (LBBB), right bundle branch block (RBBB), ventricular premature contractions (VPC) and atrial premature contractions (APC) comes under abnormal heart beats.

Este artículo presenta los resultados obtenidos al desarrollar una metodología para la detección de 5 tipos de latidos cardiacos (Normal (N), Bloqueo de Rama Derecha (RBBB), Bloqueo de Rama Izquierda (LBBB), Contracción Auricular Prematura (APC) y Contracción Ventricular Prematura (PVC)) utilizando la transformada por paquetes Wavelet de manera no adaptativa en la extracción de características de las señales cardiacas, empleando la función Shanon para cálculo de la entropía y adicionando una fase de identificación de nodos por cada tipo de señal cardiaca en el árbol Wavelet. La utilización de la transformada por paquetes Wavelet permite acceder a información obtenida de la descomposición tanto de baja como de alta frecuencia proporcionando un análisis más integral que el logrado con la transformada Wavelet discreta. Se evaluaron Wavelets madre de las familias Daubechies, Symlet 5 y Biortogonal inversa; que fueron resultado de una investigación previa en que se identificaron las Wavelet madre que mayor entropía presentaban con las señales cardiacas. Con la modalidad no adaptativa se reduce el costo computacional al utilizar los paquetes Wavelet, coste que representa la mayor desventaja de esta transformada, dando validez a la investigación realizada. Para la clasificación de los patrones cardiacos se emplearon las máquinas de soporte vectorial y el perceptrón multicapa. Con las máquinas de soporte vectorial empleando kernel de función de base radial, se logró un error de clasificación del 2,57 %.

Background/Objectives: COVID-19 is associated with multiple systemic effects, including cardiovascular complications. However, its interplay with cardiac conduction abnormalities remains underexplored. We compared the clinical profile of COVID-19 patients with pre-existing left bundle branch block (LBBB) or right bundle branch block (RBBB) at hospital admission. Methods: This study included 100 COVID-19 patients with antecedent BBB (50 LBBB, 50 RBBB). Critical cardiometabolic, renal, hematological, and inflammatory markers were measured. Logistic regression was used to identify key predictors differentiating COVID-19 patients with LBBB and RBBB. Spearman's correlations were applied to assess intra-strata associations for these variables. Results: COVID-19 patients with LBBB patients were significantly more likely to display lower systolic blood pressure (p = 0.012) but greater left atrial size (p = 0.008), left ventricular diameter (p = 0.001), and interventricular septal thickness (p = 0.023). Hematological and inflammatory markers differed, with LBBB patients being prone to exhibit higher red cell distribution width (p = 0.005), lymphocyte count (p < 0.001), neutrophil count (p = 0.045), and C-reactive protein (p < 0.001). This group also tended to show lower erythrocyte sedimentation rate (p = 0.013) and glycated hemoglobin (p = 0.045) but higher random glucose (p = 0.014). Absolute lymphocyte count, C-reactive protein, and left ventricular diameter were the most robust predictors distinguishing LBBB from RBBB. Significant associations were found exclusively for LBBB, all of them being weak. These predominantly negative relationships indicated an inflammatory origin, and most of them occurred for lymphocyte count. Conclusions: COVID-19 patients with LBBB and RBBB present distinct clinical profiles at hospital admission. The former group demonstrates a more adverse baseline clinical profile, particularly in terms of cardiac and inflammatory markers. These findings suggest that pre-existing BBB type may influence disease progression, potentially helping in risk stratification for COVID-19 patients.

Arrhythmia is caused by improper and irregular sinus rhythm or heartbeats. In order to diagnose cardiac arrhyth- mia, electrocardiogram (ECG) beat classification and analysis is very necessary. The efficiency and accuracy of any classification model highly depends on selecting the most relevant features. The aim of this study is to classify different arrhythmic beats with a reduced set of relevant-only ECG features. To optimize the ECG feature selection process and increase the classification accuracy, a Mahalanobis-Taguchi System (MTS) based classifica- tion and analysis scheme is proposed. MTS is a multi-dimensional pattern recognition system which dynamically selects important features for further analysis. Arrhythmia can occur at any time and thus requires proper and continuous monitoring of the patient to reduce sudden heart attacks. The proposed MTS- based classification scheme is integrated with a Wireless Body Area Network (WBAN) for pervasive monitoring. The proposed scheme is analyzed and compared with a state-of-the-art scheme in terms of sensitivity, specificity, and accuracy. The results show that the proposed scheme performs significantly better than the other scheme by achieving high sensitivity, specificity, and classification accuracy for different arrhythmic heartbeats i.e., Left Bundle Branch Block (LBBB), Premature Ventricular Contraction (PVC), Right Bundle Branch Block (RBBB), and Atrial Premature Contraction (APC).

Premature ventricular contraction (PVC) is one of the most serious arrhythmias. Without early diagnosis and proper treatment, PVC can result in significant complications. In this paper, a novel feature extraction method based on a sparse auto-encoder (SAE) and softmax regression (SR) classifier was used to differentiate PVCs from other common Non-PVC rhythms, including normal sinus (N), left bundle branch block (LBBB), right bundle branch block (RBBB), atrial premature contraction (APC), and paced beat (PB) rhythms. The proposed method was analyzed using 40 ECG records obtained from the MIT-BIH Arrhythmia Database. The proposed method exhibited an overall accuracy of 99.4%, with a PVC recognition sensitivity and positive predictability of 97.9% and 91.8%, respectively.

The study investigates the prognostic impact of right bundle branch block (RBBB) and left bundle branch block (LBBB) in patients with cardiogenic shock (CS) compared with no bundle branch block (BBB). In patients with heart failure, existence of RBBB and LBBB has influence on prognosis. Prospective registry-study. ICU of a tertiary academic hospital in Germany. Adult patients with CS. None. Consecutive patients with CS were included. The prognostic impact of RBBB and LBBB on 30-day all-cause mortality was tested within the entire cohort and in the subgroup of CS patients with cardiac arrest at admission. The final study cohort comprised 248 patients. Patients with RBBB showed the highest 30-day all-cause mortality followed by LBBB and no BBB (72.5% vs. 52.9% vs. 50.0%; log-rank p = 0.015). These findings were consistent even after solely including CS patients with cardiac arrest (90.0% vs. 73.3% vs. 62.2%; log-rank p = 0.008). After adjustment for lactate, norepinephrine, troponin I, Acute Physiology Score, Society of Cardiovascular Angiography & Interventions shock stage, and heart rate in a multivariable Cox regression analysis, RBBB still revealed a negative impact on 30-day all-cause mortality (hazard ratio [HR], 1.807; 95% CI, 1.107-2.947; p = 0.018), whereas LBBB was not associated with 30-day all-cause mortality. In this multivariable Cox regression model lactate (HR, 1.065; 95% CI, 1.018-1.115; p = 0.006), troponin I (HR, 1.003; 95% CI, 1.001-1.005; p = 0.001), and Acute Physiology Score (HR, 1.033; 95% CI, 1.001-1.066; p = 0.041) were as well associated with 30-day all-cause mortality. Finally, no association of RBBB was found with the incidence of liver or severe renal failure. Besides the Acute Physiology Score, lactate, and troponin levels, RBBB was associated with an increased 30-day all-cause mortality in consecutive CS patients with and without cardiac arrest, whereas LBBB showed no prognostic impact.

Millions of people around the world are affected by arrhythmias, which are abnormal activities of the functioning of the heart. Most arrhythmias are harmful to the heart and can suddenly become life-threatening. The electrocardiogram (ECG) is an important non-invasive tool in cardiology for the diagnosis of arrhythmias. This work proposes a computer-aided diagnosis (CAD) system to automatically classify different types of arrhythmias from ECG signals. First, the auto-encoder convolutional network (ACN) model is used, which is based on a one-dimensional convolutional neural network (1D-CNN)that automatically learns the best features from the raw ECG signals. After that, the support vector machine (SVM) classifier is applied to the features learned by the ACN model to improve the detection of arrhythmic beats. This classifier detects four different types of arrhythmias, namely the left bundle branch block (LBBB), right bundle branch block (RBBB), paced beat (PB), and premature ventricular contractions (PVC), along with the normal sinus rhythms (NSR). Among these arrhythmias, PVC is particularly a dangerous type of heartbeat in ECG signals. The performance of the model is measured in terms of accuracy, sensitivity, and precision using a tenfold cross-validation strategy on the MIT-BIH arrhythmia database. The obtained overall accuracy of the SVM classifier was 98.84%. The result of this model is portrayed as a better performance than in other literary works. Thus, this approach may also help in further clinical studies of cardiac cases.

The most effective method used for the diagnosis of heart diseases is the Electrocardiogram (ECG). The shape of the ECG signal and the time interval between its various components gives useful details about any underlying heart disease. Any dysfunction of the heart is called as cardiac arrhythmia. The electrical impulses of the heart are blocked due to the cardiac arrhythmia called Bundle Branch Block (BBB) which can be observed as an irregular ECG wave. The BBB beats can indicate serious heart disease. The precise and quick detection of cardiac arrhythmias from the ECG signal can save lives and can also reduce the diagnostics cost. This study presents a machine learning technique for the automatic detection of BBB. In this method both morphological and statistical features were calculated from the ECG signals available in the standard MIT BIH database to classify them as normal, Left Bundle Branch Block (LBBB) and Right Bundle Branch Block (RBBB). ECG records in the MIT- BIH arrhythmia database containing Normal sinus rhythm, RBBB, and LBBB were used in the study. The suitability of the features extracted was evaluated using three classifiers, support vector machine, k-nearest neighbours and linear discriminant analysis. The accuracy of the technique is highly promising for all the three classifiers with k-nearest neighbours giving the highest accuracy of 98.2%. Since the ECG waveforms of patients with the same cardiac disorder is similar in shape, the proposed method is subject independent. The proposed technique is thus a reliable and simple method involving less computational complexity for the automatic detection of bundle branch block. This system can reduce the effort of cardiologists thereby enabling them to concentrate more on treatment of the patients.

The features of ventricular arrhythmia in coronary spasm induced by ergonovine were examined to elucidate the characteristics of ventricular arrhythmia in acute myocardial ischemia. The coupling intervals and configurations of the QRS wave of the premature ventricular beats (PVBs) were extremely variable, even in the same individual and at the closer time phases. The prematurity index (R-R'/QT) of PVBs with the right bundle branch block (RBBB) pattern was significantly smaller than with the left bundle branch block (LBBB) pattern. The vulnerability index (R-RX QT/R-R') of PVBs with the RBBB pattern was significantly larger than with the LBBB pattern. The incidence of PVBs with the RBBB pattern (50.4%) was slightly larger than with the LBBB pattern (40.5%). PVBs deteriorated into ventricular tachycardias (VTs) in 29 (24.8%) of 117 cases with the RBBB pattern and in 11 (11.7%) out of 94 cases with the LBBB pattern. Thus, PVBs with the RBBB pattern were more likely to progress to VT than with the LBBB pattern. But it is difficult to assess the clinical significance of PVBs from their QRS morphology alone, because 11 (27.5%) of 40 cases of VT were initiated by PVBs with the LBBB pattern.

Comparison of Left Ventricular Torsion and Strain With Biventricular Pacing in Patients With Underlying Right Bundle Branch Block Versus Those With Left Bundle Branch Block

Background: Here, we present a 2D activation model evaluating the activation of specific myocardium segments (MS). Activation times of specific MS can help to identify conduction disturbances which may result in more targeted treatment of patients.Methods: 12-lead ECG signal was measured with a 5kHz sampling frequency. A total of 10 left bundle branch block (LBBB) and 5 right bundle branch block (RBBB) recordings were analyzed. The analysis includes the following steps: 1) QRS complexes detection and morphology clustering, 2) averaging of 150-1000 Hz envelopes of QRS complexes with dominant morphology, 3) a genetic algorithm (GA) produces artificial envelopes from (initially random) timing of MS activation. The task for GA is to produce envelopes the most similar to measured; then final timing reflects real activation of MS.Results: Presented model determined activation of left ventricular (LV) segments before right ventricular (RV) segments in all RBBB patients (mean 61.4 ± 13.7 ms) and activation of RV segments before the LV segments in all LBBB patients (mean 87.8 ± 15.8 ms). Computed activation of MS corresponds to the expected activation.Conclusion: We introduced a new method determining activation times of MS; this is achieved non-invasively using only ultra-high-frequency precordial ECG signal.