Scholarlink

<p>Myocardial Infarction (MI), commonly known as heart attack, is still one of the most frequent and devastating cardiovascular conditions in the world, accounting for a huge percentage of death rates. MI mortality rates are still at concerning levels, placing a significant strain on healthcare systems throughout the globe. Improving patient outcomes and initiating timely medical treatments depend on the timely and accurate diagnosis of myocardial infarction (MI). Electrocardiography (ECG) is a vital diagnostic technique in cardiology that records electrical activity to provide a non-invasive way to evaluate the heart’s health. ECG signals can help detect anomalies suggestive of MI and offer important insights into heart function. In this work, we present a new method for MI diagnosis based on an ensemble of convolutional neural networks (CNNs) trained on ECG signal data. With the aid of ensemble learning and deep learning, our model can accurately identify MI. With a 99.12% accuracy rate, 99.71% specificity, and 99.11% precision, our proposed approach demonstrates encouraging outcomes in successfully detecting MI patients and even patients with a history of MI. This CNN-Ensemble model is pertinent because it has the potential to give medical practitioners a solid and trustworthy tool for the rapid and accurate diagnosis of MI. </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

Adaptive filter-enabled electrocardiogram signals-based myocardial infarction detection using convolutional neural network ensemble model

  • 期刊论文

  • 发表日期：

    2025年6月30日

查看全文
  • Sourabh Shastri,
  • Sachin Kumar,
  • Vibhakar Mansotra

摘要

Myocardial Infarction (MI), commonly known as heart attack, is still one of the most frequent and devastating cardiovascular conditions in the world, accounting for a huge percentage of death rates. MI mortality rates are still at concerning levels, placing a significant strain on healthcare systems throughout the globe. Improving patient outcomes and initiating timely medical treatments depend on the timely and accurate diagnosis of myocardial infarction (MI). Electrocardiography (ECG) is a vital diagnostic technique in cardiology that records electrical activity to provide a non-invasive way to evaluate the heart’s health. ECG signals can help detect anomalies suggestive of MI and offer important insights into heart function. In this work, we present a new method for MI diagnosis based on an ensemble of convolutional neural networks (CNNs) trained on ECG signal data. With the aid of ensemble learning and deep learning, our model can accurately identify MI. With a 99.12% accuracy rate, 99.71% specificity, and 99.11% precision, our proposed approach demonstrates encouraging outcomes in successfully detecting MI patients and even patients with a history of MI. This CNN-Ensemble model is pertinent because it has the potential to give medical practitioners a solid and trustworthy tool for the rapid and accurate diagnosis of MI.