Scholarlink

<p>Early prevention of myocardial infarction (MI) aggravation and associated cardiac health consequences can be achieved with computerized identification and localization of MI. The segregation process within and across leads has not been thoroughly examined, even though several techniques have been used based on the deep learning covered by the published papers. The optimized Resnet-V2 is introduced for the detection of MI from the images to address the problem. The suggested approach combines Resnet-V2 with the Improved Pufferfish Optimization Algorithm (IPOA). Levy Fight (LF) and the Pufferfish Optimization Algorithm (POA) are combined to create the IPOA.LF is used in the POA to improve the updating procedure. The dataset was first gathered via internet sources. Subsequently, the images are processed to eliminate extraneous information using the Gaussian filter. We apply the 2D-convolutional neural network to segment the needed components. Ultimately, the suggested classifier is applied to identify MI in the pictures. The suggested approach is put into practice in MATLAB, and statistical measures of accuracy, sensitivity, specificity, F_Measure, precision, AUC, ROC, and Kappa are used to assess performances. Furthermore, it is contrasted with traditional methods like CNN-Particle Swarm Optimization (CNN-PSO), CNN-Genetic Algorithm (CNN-GA), and Convolutional Neural Network (CNN), in that order.</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 hybrid approach for myocardial infarction detection using improved pufferfish optimization algorithm with Resnet-V2

  • 期刊论文

  • 发表日期：

    2026年1月18日

查看全文
  • Dr. A. Franklin Alex Joseph,
  • Dr. N. Narasimhulu,
  • Perumal Pitchandi

摘要

Early prevention of myocardial infarction (MI) aggravation and associated cardiac health consequences can be achieved with computerized identification and localization of MI. The segregation process within and across leads has not been thoroughly examined, even though several techniques have been used based on the deep learning covered by the published papers. The optimized Resnet-V2 is introduced for the detection of MI from the images to address the problem. The suggested approach combines Resnet-V2 with the Improved Pufferfish Optimization Algorithm (IPOA). Levy Fight (LF) and the Pufferfish Optimization Algorithm (POA) are combined to create the IPOA.LF is used in the POA to improve the updating procedure. The dataset was first gathered via internet sources. Subsequently, the images are processed to eliminate extraneous information using the Gaussian filter. We apply the 2D-convolutional neural network to segment the needed components. Ultimately, the suggested classifier is applied to identify MI in the pictures. The suggested approach is put into practice in MATLAB, and statistical measures of accuracy, sensitivity, specificity, F_Measure, precision, AUC, ROC, and Kappa are used to assess performances. Furthermore, it is contrasted with traditional methods like CNN-Particle Swarm Optimization (CNN-PSO), CNN-Genetic Algorithm (CNN-GA), and Convolutional Neural Network (CNN), in that order.