ECG feature extraction using principal component analysis for studying the effect of diabetes

Feasibility Analysis of ECG-Based pH Estimation for Asphyxia Detection in Neonates

Birth asphyxia is a potential cause of death that is also associated with acute and chronic morbidities. The traditional and immediate approach for monitoring birth asphyxia (i.e., arterial blood gas analysis) is highly invasive and intermittent. Additionally, alternative noninvasive approaches such as pulse oximeters can be problematic, due to the possibility of false and erroneous measurements. Therefore, further research is needed to explore alternative noninvasive and accurate monitoring methods for asphyxiated neonates. This study aims to investigate the prominent ECG features based on pH estimation that could potentially be used to explore the noninvasive, accurate, and continuous monitoring of asphyxiated neonates. The dataset used contained 274 segments of ECG and pH values recorded simultaneously. After preprocessing the data, principal component analysis and the Pan-Tompkins algorithm were used for each segment to determine the most significant ECG cycle and to compute the ECG features. Descriptive statistics were performed to describe the main properties of the processed dataset. A Kruskal-Wallis nonparametric test was then used to analyze differences between the asphyxiated and non-asphyxiated groups. Finally, a Dunn-Šidák post hoc test was used for individual comparison among the mean ranks of all groups. The findings of this study showed that ECG features (T/QRS, T Amplitude, Tslope, Tslope/T, Tslope/|T|, HR, QT, and QTc) based on pH estimation differed significantly (p < 0.05) in asphyxiated neonates. All these key ECG features were also found to be significantly different between the two groups.

Multi-Domain Variational Autoencoders for Combined Modeling of MRI-Based Biventricular Anatomy and ECG-Based Cardiac Electrophysiology

Human cardiac function is characterized by a complex interplay of mechanical deformation and electrophysiological conduction. Similar to the underlying cardiac anatomy, these interconnected physiological patterns vary considerably across the human population with important implications for the effectiveness of clinical decision-making and the accuracy of computerized heart models. While many previous works have investigated this variability separately for either cardiac anatomy or physiology, this work aims to combine both aspects in a single data-driven approach and capture their intricate interdependencies in a multi-domain setting. To this end, we propose a novel multi-domain Variational Autoencoder (VAE) network to capture combined Electrocardiogram (ECG) and Magnetic Resonance Imaging (MRI)-based 3D anatomy information in a single model. Each VAE branch is specifically designed to address the particular challenges of the respective input domain, enabling efficient encoding, reconstruction, and synthesis of multi-domain cardiac signals. Our method achieves high reconstruction accuracy on a United Kingdom Biobank dataset, with Chamfer Distances between reconstructed and input anatomies below the underlying image resolution and ECG reconstructions outperforming multiple single-domain benchmarks by a considerable margin. The proposed VAE is capable of generating realistic virtual populations of arbitrary size with good alignment in clinical metrics between the synthesized and gold standard anatomies and Maximum Mean Discrepancy (MMD) scores of generated ECGs below those of comparable single-domain approaches. Furthermore, we observe the latent space of our VAE to be highly interpretable with separate components encoding different aspects of anatomical and ECG variability. Finally, we demonstrate that the combined anatomy and ECG representation improves the performance in a cardiac disease classification task by 3.9% in terms of Area Under the Receiver Operating Characteristic (AUROC) curve over the best corresponding single-domain modeling approach.

A cascaded classifier for multi-lead ECG based on feature fusion

BACKGROUND AND OBJECTIVE

Electrocardiogram (ECG) is an important diagnostic tool for the diagnosis of heart disorders. Useful features and well-designed classification method are crucial for automatic diagnosis. However, most of the contributions were in single lead or two-lead ECG signal and only features from single lead were used to classify the ECG beats. In this paper, a cascaded classification system is proposed to extract features and classify heartbeats in order to improve the performance of ECG beat classification via multi-lead ECG.

METHODS

In contrast with most of the literatures, ten signal features were chosen and run on each of the 12 leads separately. Based on these features, we developed a novel feature fusion method combining information from all available leads, and then implemented a cascaded classifier utilizing random forest (RF) and multilayer perceptron (MLP). Besides, in order to reduce the feature space dimension, principal component analysis (PCA) was applied in the method.

MATERIALS

Four open source databases including MIT-BIH Arrythmia Database, QT Database, MIT-BIH Supraventricular Arrhythmia Database and St. Petersburg Institute of Cardiological Technics 12-lead Arrhythmia Database (INCART Database) were used in this work. These four databases are different in classes of beats, volume of dataset, number of individual volunteers. Except INCART database, in which the recording format is 12-lead, the other three databases consist of 2-lead recordings. Above all, they all have annotations for every single beat including the type of each beat.

CONCLUSIONS

Extensive experimental results shown that the average accuracy achieved 99.3%, 99.8%, 97.6% and 99.6% on four databases respectively. Compared with most state-of-the-art methods, our work has better performance and strong generalization capability.

Feature extraction of ECG signal

This paper deals with new approaches to analyse electrocardiogram (ECG) signals for extracting useful diagnostic features. Initially, elimination of different types of noise is carried out using maximal overlap discrete wavelet transform (MODWT) and universal thresholding. Next, R-peak fiducial points are detected from these noise free ECG signals using discrete wavelet transform along with thresholding. Then, extraction of other features, viz., Q waves, S waves, P waves, T waves, P wave onset and offset points, T wave onset and offset points, QRS onset and offset points are identified using some rule based algorithms. Eventually, other important features are computed using the above extracted features. The software developed for this purpose has been validated by extensive testing of ECG signals acquired from the MIT-BIH database. The resulting signals and tabular results illustrate the performance of the proposed method. The sensitivity, predictivity and error of beat detection are 99.98%, 99.97% and 0.05%, respectively. The performance of the proposed beat detection method is compared to other existing techniques, which shows that the proposed method is superior to other methods.

Combining Low-dimensional Wavelet Features and Support Vector Machine for Arrhythmia Beat Classification

Automatic feature extraction and classification are two main tasks in abnormal ECG beat recognition. Feature extraction is an important prerequisite prior to classification since it provides the classifier with input features, and the performance of classifier depends significantly on the quality of these features. This study develops an effective method to extract low-dimensional ECG beat feature vectors. It employs wavelet multi-resolution analysis to extract time-frequency domain features and then applies principle component analysis to reduce the dimension of the feature vector. In classification, 12-element feature vectors characterizing six types of beats are used as inputs for one-versus-one support vector machine, which is conducted in form of 10-fold cross validation with beat-based and record-based training schemes. Tested upon a total of 107049 beats from MIT-BIH arrhythmia database, our method has achieved average sensitivity, specificity and accuracy of 99.09%, 99.82% and 99.70%, respectively, using the beat-based training scheme, and 44.40%, 88.88% and 81.47%, respectively, using the record-based training scheme.