• Corpus ID: 211010663

Electrocardiogram Generation and Feature Extraction Using a Variational Autoencoder

  title={Electrocardiogram Generation and Feature Extraction Using a Variational Autoencoder},
  author={V. V. Kuznetsov and V. A. Moskalenko and Nikolai Yu. Zolotykh},
We propose a method for generating an electrocardiogram (ECG) signal for one cardiac cycle using a variational autoencoder. Using this method we extracted a vector of new 25 features, which in many cases can be interpreted. The generated ECG has quite natural appearance. The low value of the Maximum Mean Discrepancy metric, 0.00383, indicates good quality of ECG generation too. The extracted new features will help to improve the quality of automatic diagnostics of cardiovascular diseases. Also… 

Figures from this paper

An Interpretable Model for ECG Data Based on Bayesian Neural Networks
It is found that the features extracted from raw ECG data using the developed framework can be used as an indication for specific heartbeat arrhythmia, and the mechanism of the model can be interpreted as a decision making procedure by weighing relative metrics just like human experts do.
Classification of Arrhythmia in Heartbeat Detection Using Deep Learning
The electrocardiogram (ECG) is one of the most widely used diagnostic instruments in medicine and healthcare. Deep learning methods have shown promise in healthcare prediction challenges involving
Applications of Machine Learning in Ambulatory ECG
The center tasks of AECG signal processing include signal preprocessing, beat detection and classification, event detection, and event prediction, which are the topics most relevant and of greatest concern to the people working in this area.


Deep Learning for ECG Segmentation
An algorithm for electrocardiogram segmentation using a UNet-like full-convolutional neural network that is adaptive to different sampling rates and generalized to various types of ECG monitors is proposed.
PGANs: Personalized Generative Adversarial Networks for ECG Synthesis to Improve Patient-Specific Deep ECG Classification
A generative model is proposed that learns to synthesize patient-specific ECG signals, which can then be used as additional training data to improve a patient- specific classifier performance.
Electrocardiogram generation with a bidirectional LSTM-CNN generative adversarial network
A generative adversarial network, which is composed of a bidirectional long short-term memory and convolutional neural network, referred as BiLSTM-CNN, to generate synthetic ECG data that agree with existing clinical data so that the features of patients with heart disease can be retained.
LU electrocardiography database: a new open-access validation tool for delineation algorithms
A new Lobachevsky University database of ECG signals that contains 200 samples of 10-second 12-lead electrocardiograms (ECG) from different subjects and has an attributed diagnosis makes LUDB a promising tool for validating ECG delineation algorithms across a broad range of signal shapes and patient diagnoses.
Synthesis of Realistic ECG using Generative Adversarial Networks
For the first time, it is both quantitatively and qualitatively demonstrate that GAN architecture can successfully generate time series signals that are not only structurally similar to the training sets but also diverse in nature across generated samples.
Finding Morphology Points of Electrocardiographic-Signal Waves Using Wavelet Analysis
We propose a new algorithm for determining the basic significant points of various electrocardiographic-signal waves taking into account information from all available leads and ensuring a similar or
Tutorial on Variational Autoencoders
This tutorial introduces the intuitions behind VAEs, explains the mathematics behind them, and describes some empirical behavior.
Auto-Encoding Variational Bayes
A stochastic variational inference and learning algorithm that scales to large datasets and, under some mild differentiability conditions, even works in the intractable case is introduced.
Principal Manifolds for Data Visualization and Dimension Reduction
The book is meant to be useful for practitioners in applied data analysis in life sciences, engineering, physics and chemistry; it will also be valuable to PhD ...