Uma Maheswari K, Valarmathi A
Department of Information Technology, University College of Engineering, BIT Campus, Anna University, Tiruchirappalli, India.
Department of Computer Applications, University College of Engineering, BIT Campus, Anna University, Tiruchirappalli, India.
Front Physiol. 2025 Jun 5;16:1563199. doi: 10.3389/fphys.2025.1563199. eCollection 2025.
Over the years, the number of people who succumbed to heart ailments has increased significantly worldwide. The World Health Organization claims that about 17 million people die each year due to heart disease. High levels of cholesterol and blood pressure are some risk factors. This technology seeks to treat these conditions before they become a problem. Through machine learning, doctors can now make more informed decisions regarding the treatment of patients. Machine learning can assist in reducing the likelihood of a cardiac event. Conventional methods for diagnosing diseases often lead to inaccurate diagnoses and take longer to complete due to human errors. In order to increase the diagnostic accuracy, an ensemble method is used. This method combines various classifiers to achieve highly accurate predictions. Due to the complexity of the task, the researchers decided to use deep learning methods to perform the heart disease classification task. H-DSAE technique utilize Deep Belief Network (DBN), Support Vector Machine (SVM), and Stacked Auto-Encoder (SAE). It was able to extract various heart image representations and achieve an accuracy of 99.2. It also had a sensitivity of 97.5, F-measure of 98.5, and precision of 98.4. The next phase of the project will focus on developing more advanced classification and features algorithms. This will help improve the efficiency of the system.
多年来，全球因心脏病死亡的人数显著增加。世界卫生组织称，每年约有1700万人死于心脏病。高胆固醇和高血压是一些风险因素。这项技术旨在在这些疾病成为问题之前进行治疗。通过机器学习，医生现在可以在治疗患者方面做出更明智的决策。机器学习有助于降低心脏事件发生的可能性。传统的疾病诊断方法往往导致诊断不准确，并且由于人为错误需要更长时间才能完成。为了提高诊断准确性，使用了一种集成方法。该方法结合了各种分类器以实现高度准确的预测。由于任务的复杂性，研究人员决定使用深度学习方法来执行心脏病分类任务。H-DSAE技术利用深度信念网络（DBN）、支持向量机（SVM）和堆叠自编码器（SAE）。它能够提取各种心脏图像特征，并达到99.2的准确率。它还具有97.5的灵敏度、98.5的F值和98.4的精确率。该项目的下一阶段将专注于开发更先进的分类和特征算法。这将有助于提高系统的效率。
