Scholarlink

<p>Efficient diagnosis of cardiovascular diseases is a key issue due to data imbalance in electrocardiogram (ECG) datasets, which can lead to bias in model predictions and hinder real-world applicability. In this paper, we introduce AugEViT, a supervised model for multi-class and multi-label classification of arrhythmia in 12-lead ECG recordings, addressing data imbalance in ECG datasets. The model uses a neural network architecture based on a Vision Transformer model and data augmentation techniques, achieving an average accuracy of 98.7% and an AUC ROC of 98% to recognize 51 distinct labels for various arrhythmia types.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A data-augmented vision transformer model for robust multi-label ECG arrhythmia classification

  • 期刊论文

  • 发表日期：

    2025年1月27日

查看全文
  • Aymen Abid,
  • Omar Cheikhrouhou

摘要

Efficient diagnosis of cardiovascular diseases is a key issue due to data imbalance in electrocardiogram (ECG) datasets, which can lead to bias in model predictions and hinder real-world applicability. In this paper, we introduce AugEViT, a supervised model for multi-class and multi-label classification of arrhythmia in 12-lead ECG recordings, addressing data imbalance in ECG datasets. The model uses a neural network architecture based on a Vision Transformer model and data augmentation techniques, achieving an average accuracy of 98.7% and an AUC ROC of 98% to recognize 51 distinct labels for various arrhythmia types.