Key result
ECG-Adversarial network outperforms standard models with 0.91 ROC-AUC for classifying cardiac conditions from mobile ECGs.
Why the study?
Much 12-lead ECG data remains in printed or image formats, and using smartphone cameras to digitize them introduces artifacts and occlusions that hinder automated analysis.
Does a domain adversarial neural network improve the diagnostic accuracy of cardiac conditions from mobile-captured 12-lead ECG images compared to standard convolutional neural networks?
Does a domain adversarial neural network improve the diagnostic accuracy of cardiac conditions from mobile-captured 12-lead ECG images compared to standard convolutional neural networks?
Absolute Event Rate: 0.91% vs 0.62%
Domain adversarial training enables highly accurate automated interpretation of 12-lead ECGs from mobile phone photos, overcoming artifacts and format variations.
Should not yet change ECG interpretation practice; extends domain adaptation methods while leaving open prospective clinical validation.
12-lead electrocardiogram (ECG) recordings can be collected in any clinic and the interpretation is performed by a clinician. Modern machine learning tools may make them automatable. However, a large fraction of 12-lead ECG data is still available in printed paper or image only and comes in various formats. To digitize the data, smartphone cameras can be used. Nevertheless, this approach may introduce various artifacts and occlusions into the obtained images. Here we overcome the challenges of automating 12-lead ECG analysis using mobile-captured images and a deep neural network that is trained using a domain adversarial approach. The net achieved an average 0.91 receiver operating characteristic curve on tested images captured by a mobile device. Assessment on image from unseen 12-lead ECG formats that the network was not trained on achieved high accuracy. We further show that the network accuracy can be improved by including a small number of unlabeled samples from unknown formats in the training data. Finally, our models also achieve high accuracy using signals as input rather than images. Using a domain adaptation approach, we successfully classified cardiac conditions on images acquired by a mobile device and showed the generalizability of the classification using various unseen image formats.
No takes yet. Share an insight, caveat, or question.
Глинер et al. (2023) studied Cardiac conditions (14 morphological and arrhythmogenic disorders) (n=118,839). ECG-Adversarial (domain adversarial neural network) vs. ECG-Vanilla (standard convolutional neural network) was evaluated on Average ROC-AUC for detection of 14 cardiac conditions on mobile device-acquired 12-lead ECG images. The ECG-Adversarial domain adaptation neural network achieved an average ROC-AUC of 0.91 for classifying 14 cardiac conditions from mobile-captured 12-lead ECG images, outperforming a standard convolutional network.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: