A Comprehensive Benchmark for Electrocardiogram Time-Series

TL;DR

This paper introduces a comprehensive benchmark for ECG time-series analysis, including new evaluation metrics and a novel model architecture, to improve understanding and performance in cardiac health diagnostics.

Contribution

It provides a detailed categorization of ECG applications, identifies limitations of existing metrics, and proposes a new evaluation metric and model architecture tailored for ECG data.

Findings

The new benchmark is comprehensive and robust.

The proposed metric improves evaluation accuracy.

The new architecture outperforms existing models.

Abstract

Electrocardiogram~(ECG), a key bioelectrical time-series signal, is crucial for assessing cardiac health and diagnosing various diseases. Given its time-series format, ECG data is often incorporated into pre-training datasets for large-scale time-series model training. However, existing studies often overlook its unique characteristics and specialized downstream applications, which differ significantly from other time-series data, leading to an incomplete understanding of its properties. In this paper, we present an in-depth investigation of ECG signals and establish a comprehensive benchmark, which includes (1) categorizing its downstream applications into four distinct evaluation tasks, (2) identifying limitations in traditional evaluation metrics for ECG analysis, and introducing a novel metric; (3) benchmarking state-of-the-art time-series models and proposing a new architecture. Extensive experiments demonstrate that our proposed benchmark is comprehensive and robust. The results validate the effectiveness of the proposed metric and model architecture, which establish a solid foundation for advancing research in ECG signal analysis.