ME-GAN: Learning Panoptic Electrocardio Representations for Multi-view ECG Synthesis Conditioned on Heart Diseases

Jintai Chen, Kuanlun Liao, Kun Wei, Haochao Ying, Danny Z Chen, Jian Wu
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:3360-3370, 2022.

Abstract

Electrocardiogram (ECG) is a widely used non-invasive diagnostic tool for heart diseases. Many studies have devised ECG analysis models (e.g., classifiers) to assist diagnosis. As an upstream task, researches have built generative models to synthesize ECG data, which are beneficial to providing training samples, privacy protection, and annotation reduction. However, previous generative methods for ECG often neither synthesized multi-view data, nor dealt with heart disease conditions. In this paper, we propose a novel disease-aware generative adversarial network for multi-view ECG synthesis called ME-GAN, which attains panoptic electrocardio representations conditioned on heart diseases and projects the representations onto multiple standard views to yield ECG signals. Since ECG manifestations of heart diseases are often localized in specific waveforms, we propose a new "mixup normalization" to inject disease information precisely into suitable locations. In addition, we propose a "view discriminator" to revert disordered ECG views into a pre-determined order, supervising the generator to obtain ECG representing correct view characteristics. Besides, a new metric, rFID, is presented to assess the quality of the synthesized ECG signals. Comprehensive experiments verify that our ME-GAN performs well on multi-view ECG signal synthesis with trusty morbid manifestations.

Cite this Paper

BibTeX
@InProceedings{pmlr-v162-chen22n, title = {{ME}-{GAN}: Learning Panoptic Electrocardio Representations for Multi-view {ECG} Synthesis Conditioned on Heart Diseases}, author = {Chen, Jintai and Liao, Kuanlun and Wei, Kun and Ying, Haochao and Chen, Danny Z and Wu, Jian}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {3360--3370}, year = {2022}, editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan}, volume = {162}, series = {Proceedings of Machine Learning Research}, month = {17--23 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v162/chen22n/chen22n.pdf}, url = {https://proceedings.mlr.press/v162/chen22n.html}, abstract = {Electrocardiogram (ECG) is a widely used non-invasive diagnostic tool for heart diseases. Many studies have devised ECG analysis models (e.g., classifiers) to assist diagnosis. As an upstream task, researches have built generative models to synthesize ECG data, which are beneficial to providing training samples, privacy protection, and annotation reduction. However, previous generative methods for ECG often neither synthesized multi-view data, nor dealt with heart disease conditions. In this paper, we propose a novel disease-aware generative adversarial network for multi-view ECG synthesis called ME-GAN, which attains panoptic electrocardio representations conditioned on heart diseases and projects the representations onto multiple standard views to yield ECG signals. Since ECG manifestations of heart diseases are often localized in specific waveforms, we propose a new "mixup normalization" to inject disease information precisely into suitable locations. In addition, we propose a "view discriminator" to revert disordered ECG views into a pre-determined order, supervising the generator to obtain ECG representing correct view characteristics. Besides, a new metric, rFID, is presented to assess the quality of the synthesized ECG signals. Comprehensive experiments verify that our ME-GAN performs well on multi-view ECG signal synthesis with trusty morbid manifestations.} }
Endnote
%0 Conference Paper %T ME-GAN: Learning Panoptic Electrocardio Representations for Multi-view ECG Synthesis Conditioned on Heart Diseases %A Jintai Chen %A Kuanlun Liao %A Kun Wei %A Haochao Ying %A Danny Z Chen %A Jian Wu %B Proceedings of the 39th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2022 %E Kamalika Chaudhuri %E Stefanie Jegelka %E Le Song %E Csaba Szepesvari %E Gang Niu %E Sivan Sabato %F pmlr-v162-chen22n %I PMLR %P 3360--3370 %U https://proceedings.mlr.press/v162/chen22n.html %V 162 %X Electrocardiogram (ECG) is a widely used non-invasive diagnostic tool for heart diseases. Many studies have devised ECG analysis models (e.g., classifiers) to assist diagnosis. As an upstream task, researches have built generative models to synthesize ECG data, which are beneficial to providing training samples, privacy protection, and annotation reduction. However, previous generative methods for ECG often neither synthesized multi-view data, nor dealt with heart disease conditions. In this paper, we propose a novel disease-aware generative adversarial network for multi-view ECG synthesis called ME-GAN, which attains panoptic electrocardio representations conditioned on heart diseases and projects the representations onto multiple standard views to yield ECG signals. Since ECG manifestations of heart diseases are often localized in specific waveforms, we propose a new "mixup normalization" to inject disease information precisely into suitable locations. In addition, we propose a "view discriminator" to revert disordered ECG views into a pre-determined order, supervising the generator to obtain ECG representing correct view characteristics. Besides, a new metric, rFID, is presented to assess the quality of the synthesized ECG signals. Comprehensive experiments verify that our ME-GAN performs well on multi-view ECG signal synthesis with trusty morbid manifestations.
APA
Chen, J., Liao, K., Wei, K., Ying, H., Chen, D.Z. & Wu, J.. (2022). ME-GAN: Learning Panoptic Electrocardio Representations for Multi-view ECG Synthesis Conditioned on Heart Diseases. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:3360-3370 Available from https://proceedings.mlr.press/v162/chen22n.html.

Related Material