Enhancing Deep Learning-based 3-lead ECG Classification with Heartbeat Counting and Demographic Data Integration

TL;DR

This paper introduces two innovative techniques—heartbeat counting via multi-task learning and demographic data integration—to significantly enhance 3-lead ECG classification, achieving performance comparable to 12-lead systems and surpassing existing methods.

Contribution

The paper presents a novel multi-task learning approach and demographic data integration to improve 3-lead ECG classification accuracy, bridging the gap with 12-lead systems.

Findings

Achieved F1 scores of 0.9796 and 0.8140 on Chapman and CPSC-2018 datasets.

Surpassed state-of-the-art ECG classification methods.

Demonstrated effectiveness of heartbeat counting and demographic data integration.

Abstract

Nowadays, an increasing number of people are being diagnosed with cardiovascular diseases (CVDs), the leading cause of death globally. The gold standard for identifying these heart problems is via electrocardiogram (ECG). The standard 12-lead ECG is widely used in clinical practice and the majority of current research. However, using a lower number of leads can make ECG more pervasive as it can be integrated with portable or wearable devices. This article introduces two novel techniques to improve the performance of the current deep learning system for 3-lead ECG classification, making it comparable with models that are trained using standard 12-lead ECG. Specifically, we propose a multi-task learning scheme in the form of the number of heartbeats regression and an effective mechanism to integrate patient demographic data into the system. With these two advancements, we got classification performance in terms of F1 scores of 0.9796 and 0.8140 on two large-scale ECG datasets, i.e., Chapman and CPSC-2018, respectively, which surpassed current state-of-the-art ECG classification methods, even those trained on 12-lead data. To encourage further development, our source code is publicly available at https://github.com/lhkhiem28/LightX3ECG.