Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Enhancing Imbalanced Electrocardiogram Classification: A Novel Approach Integrating Data Augmentation through Wavelet Transform and Interclass Fusion

Haijian Shao, Wei Liu, Xing Deng, Daze Lu

TL;DR

This work tackles the dual challenges of imbalanced and noisy ECG data for automated arrhythmia classification. It introduces a wavelet-based interclass fusion framework that builds balanced training and test feature libraries, augmented by PCA-based data cleansing and frequency-domain averaging, and evaluates across multiple deep-learning architectures on the CPSC2018 dataset. The approach yields strong per-class performance, notably improving results for minority classes and maintaining robustness under noise, with Inception achieving particularly notable gains. The method demonstrates practical potential for enhancing ECG diagnostics and provides open-source code to facilitate reproducible, cross-dataset application in clinical settings.

Abstract

Imbalanced electrocardiogram (ECG) data hampers the efficacy and resilience of algorithms in the automated processing and interpretation of cardiovascular diagnostic information, which in turn impedes deep learning-based ECG classification. Notably, certain cardiac conditions that are infrequently encountered are disproportionately underrepresented in these datasets. Although algorithmic generation and oversampling of specific ECG signal types can mitigate class skew, there is a lack of consensus regarding the effectiveness of such techniques in ECG classification. Furthermore, the methodologies and scenarios of ECG acquisition introduce noise, further complicating the processing of ECG data. This paper presents a significantly enhanced ECG classifier that simultaneously addresses both class imbalance and noise-related challenges in ECG analysis, as observed in the CPSC 2018 dataset. Specifically, we propose the application of feature fusion based on the wavelet transform, with a focus on wavelet transform-based interclass fusion, to generate the training feature library and the test set feature library. Subsequently, the original training and test data are amalgamated with their respective feature databases, resulting in more balanced training and test datasets. Employing this approach, our ECG model achieves recognition accuracies of up to 99%, 98%, 97%, 98%, 96%, 92%, and 93% for Normal, AF, I-AVB, LBBB, RBBB, PAC, PVC, STD, and STE, respectively. Furthermore, the average recognition accuracy for these categories ranges between 92\% and 98\%. Notably, our proposed data fusion methodology surpasses any known algorithms in terms of ECG classification accuracy in the CPSC 2018 dataset.

Enhancing Imbalanced Electrocardiogram Classification: A Novel Approach Integrating Data Augmentation through Wavelet Transform and Interclass Fusion

TL;DR

This work tackles the dual challenges of imbalanced and noisy ECG data for automated arrhythmia classification. It introduces a wavelet-based interclass fusion framework that builds balanced training and test feature libraries, augmented by PCA-based data cleansing and frequency-domain averaging, and evaluates across multiple deep-learning architectures on the CPSC2018 dataset. The approach yields strong per-class performance, notably improving results for minority classes and maintaining robustness under noise, with Inception achieving particularly notable gains. The method demonstrates practical potential for enhancing ECG diagnostics and provides open-source code to facilitate reproducible, cross-dataset application in clinical settings.

Abstract

Imbalanced electrocardiogram (ECG) data hampers the efficacy and resilience of algorithms in the automated processing and interpretation of cardiovascular diagnostic information, which in turn impedes deep learning-based ECG classification. Notably, certain cardiac conditions that are infrequently encountered are disproportionately underrepresented in these datasets. Although algorithmic generation and oversampling of specific ECG signal types can mitigate class skew, there is a lack of consensus regarding the effectiveness of such techniques in ECG classification. Furthermore, the methodologies and scenarios of ECG acquisition introduce noise, further complicating the processing of ECG data. This paper presents a significantly enhanced ECG classifier that simultaneously addresses both class imbalance and noise-related challenges in ECG analysis, as observed in the CPSC 2018 dataset. Specifically, we propose the application of feature fusion based on the wavelet transform, with a focus on wavelet transform-based interclass fusion, to generate the training feature library and the test set feature library. Subsequently, the original training and test data are amalgamated with their respective feature databases, resulting in more balanced training and test datasets. Employing this approach, our ECG model achieves recognition accuracies of up to 99%, 98%, 97%, 98%, 96%, 92%, and 93% for Normal, AF, I-AVB, LBBB, RBBB, PAC, PVC, STD, and STE, respectively. Furthermore, the average recognition accuracy for these categories ranges between 92\% and 98\%. Notably, our proposed data fusion methodology surpasses any known algorithms in terms of ECG classification accuracy in the CPSC 2018 dataset.
Paper Structure (18 sections, 8 equations, 9 figures, 3 tables, 1 algorithm)

This paper contains 18 sections, 8 equations, 9 figures, 3 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Class-imbalance-related approaches for ECG analysis
  3. Noise robustness-related approaches for ECG analysis
  4. Deep Learning Classification Approach for Class-imbalanced ECG Signals
  5. Strategies to reduce class imbalance
  6. Wavelet-based data cleaning and fusion method
  7. Wavelet-based feature fusion method
  8. Dataset for performance evaluation
  9. ECG classification methods based on deep learning
  10. Experiments
  11. Experiment results of the class-imbalance ECG
  12. Model robustness performance verification and result comparison
  13. Experimental Results
  14. Conclusions
  15. Acknowledgements
  16. ...and 3 more sections

Figures (9)

  • Figure 1: The classes in CPSC 2018 dataset after pre-processing
  • Figure 2: The scheme that handles class imbalance
  • Figure 3: Schematic diagram of the fusion rules between categories of unbalanced data
  • Figure 4: The processing diagram of new training set generation
  • Figure 5: The processing diagram of new validation set generation
  • ...and 4 more figures