One Dimensional CNN ECG Mamba for Multilabel Abnormality Classification in 12 Lead ECG

TL;DR

The paper tackles multilabel abnormality detection from 12-lead ECGs, where long sequential signals challenge conventional CNNs and transformers. It introduces 1DCNN-ECG-Mamba, a hybrid model that combines a 1D CNN frontend with a bidirectional Vision Mamba encoder built on a selective state space model, achieving linear-time sequence processing with $O(n)$ complexity. The approach yields state-of-the-art AUPRC and AUROC on PhysioNet CinC 2020/2021 benchmarks, outperforming transformer-, residual-, and prior Mamba-based methods, demonstrating strong potential for clinical deployment. This work advances reliable ECG classification and supports scalable, telemedicine-friendly diagnostic tools in diverse healthcare settings.

Abstract

Accurate detection of cardiac abnormalities from electrocardiogram recordings is regarded as essential for clinical diagnostics and decision support. Traditional deep learning models such as residual networks and transformer architectures have been applied successfully to this task, but their performance has been limited when long sequential signals are processed. Recently, state space models have been introduced as an efficient alternative. In this study, a hybrid framework named One Dimensional Convolutional Neural Network Electrocardiogram Mamba is introduced, in which convolutional feature extraction is combined with Mamba, a selective state space model designed for effective sequence modeling. The model is built upon Vision Mamba, a bidirectional variant through which the representation of temporal dependencies in electrocardiogram data is enhanced. Comprehensive experiments on the PhysioNet Computing in Cardiology Challenges of 2020 and 2021 were conducted, and superior performance compared with existing methods was achieved. Specifically, the proposed model achieved substantially higher AUPRC and AUROC scores than those reported by the best previously published algorithms on twelve lead electrocardiograms. These results demonstrate the potential of Mamba-based architectures to advance reliable ECG classification. This capability supports early diagnosis and personalized treatment, while enhancing accessibility in telemedicine and resource-constrained healthcare systems.

Figures (2)

• Figure 1: The 1DCNN-ECG-Mamba architecture.
• Figure 2: Learning Rate Schedule with Warm-up and Cosine Annealing