SILICONet : A Siamese Lead Invariant Convolutional Network for Ventricular Heartbeat Detection in Electrocardiograms (ECG)

Pierre Gabriel Aublin, Jacques Felblinger, Julien Oster

Keywords: health applications, time-series, self-supervision, biomedical signal processing, classification, ecg

TL;DR: We proposed a siamese framework for the pretraining of a convolutional neural network on the Computing in Cardiology 2021 dataset therefore making it invariant to ECG lead configuration changes.

Abstract:

Pretraining deep learning models on a large corpus of unlabelled data using self supervised learning approaches can be an efficient mitigation strategy to deal with the lack of annotated data. We proposed to use a siamese framework for the pretraining of a convolutional neural network on the Computing in Cardiology 2021 dataset therefore making it invariant to ECG lead configuration changes.
The obtained representation was then trained and tested on a heartbeat classification task on the MIT BIH Arrhythmia database, and on an external independent set, namely the INCART database.
The proposed model reached a median F1 score of 0.89 on the MIT BIH Arrhythmia database comparable to the 0.90 F1 score obtained without pretraining. However, the pretrained model obtained a median F1 score of 0.74 on average over the different leads, compared to 0.53 the model without pretraining.
The proposed pretraining approach, leveraging the availability of relatively large database of un-(or weakly)annotated ECG data, allows for the training of more generalisable, lead-agnostic, heartbeat classification models. Such an approach would ensure avoiding overfitting complex deep learning models on the small MIT-BIH arrhythmia database.

Anonymity: We agree to keep the submission (including any supplements and/or code) anonymous.

Formatting: We confirm that we read and complied with the author's instructions.