Scholarlink

This study addresses the challenges of early atrial fibrillation (AF) detection by introducing a multimodal dynamic ECG data fusion approach powered by a Transformer-based analysis model. Leveraging publicly available MIT-BIH datasets and private noisy data, we constructed a robust preprocessing pipeline, ensuring effective QRS wave detection and noise reduction. The integration of RR intervals and heart rate graphs into a Transformer architecture enables precise temporal and multidimensional ECG signal analysis. Experimental results demonstrate the proposed method’s superior performance, achieving significant improvements in key evaluation metrics compared to existing techniques.

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

Multimodal Dynamic ECG Fusion for Early Atrial Fibrillation Detection Using Transformer-Based Analysis

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Yicheng Liu,
  • Jijun Tong,
  • Shudong Xia,
  • Yiran Wang

摘要

This study addresses the challenges of early atrial fibrillation (AF) detection by introducing a multimodal dynamic ECG data fusion approach powered by a Transformer-based analysis model. Leveraging publicly available MIT-BIH datasets and private noisy data, we constructed a robust preprocessing pipeline, ensuring effective QRS wave detection and noise reduction. The integration of RR intervals and heart rate graphs into a Transformer architecture enables precise temporal and multidimensional ECG signal analysis. Experimental results demonstrate the proposed method’s superior performance, achieving significant improvements in key evaluation metrics compared to existing techniques.