Scholarlink

Prompt and accurate detection of cardiac abnormalities is vital for enhancing analysis and patient care. Traditional ECG analysis relies upon on guide interpretation, that’s time-extensive and liable to errors. To deal with this, we present a deep learning-based totally technique that classifies ECG signals into five categories (N, S, V, F, and Q), each representing a particular heart circumstance. The ECG pictures are preprocessed via scaling, normalization, and augmentation to beautify model overall performance. We compare three superior architectures—VGG-16, VGG-19, and ResNet-152—with VGG-19 achieving the very best accuracy of 95.88%. The outcomes show that deep studying gives a faster, more reliable alternative to guide ECG evaluation, assisting early analysis and greater effective medical selection-making in cardiology.

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

ECG-Driven Cardiac Condition Detection Using Deep Neural Networks

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Dega Balakotaiah,
  • P. Venkata Rajulu,
  • Thirumala Devi Kommalapati,
  • Boyapati Bala Vamsi,
  • Bommu Jagadeesh Reddy

摘要

Prompt and accurate detection of cardiac abnormalities is vital for enhancing analysis and patient care. Traditional ECG analysis relies upon on guide interpretation, that’s time-extensive and liable to errors. To deal with this, we present a deep learning-based totally technique that classifies ECG signals into five categories (N, S, V, F, and Q), each representing a particular heart circumstance. The ECG pictures are preprocessed via scaling, normalization, and augmentation to beautify model overall performance. We compare three superior architectures—VGG-16, VGG-19, and ResNet-152—with VGG-19 achieving the very best accuracy of 95.88%. The outcomes show that deep studying gives a faster, more reliable alternative to guide ECG evaluation, assisting early analysis and greater effective medical selection-making in cardiology.