Scholarlink

The Anticipation of heart diseases has garnered considerable scholarly interest due to its potential to preserve life through the facilitation of prompt identification and intervention. The aim of this effort is to create an optimal Bidirectional Gated Recurrent Unit (Bi-GRU) model that accurately predicts cardiac illness using data mining approaches. We provide the Grey Wolf Optimization (GWO) method, which effectively adjusts the Bi-GRU model’s hyper parameters to escalate performance. The dataset’s objective variable indicates the presence or absence of cardiac disease. Age, sex, type of chest pain (CP), Cholesterol (chol), fasting blood sugar (fbs), resting blood pressure (trestbps), and other clinical characteristics are also included. The Bi-GRU architecture is utilised to identify temporal correlations in the patient data after feature extraction and pre-processing, while GWO optimises factors including batch size, learning rate, and hidden unit count. The optimized Bi-GRU model outperforms conventional models in terms of prediction and reliability, according to experimental data. Ultimately, by showing how bio-inspired optimization and advanced DL architectures may be applied to escalate the prediction of cardiac illness, our approach enhances the domains of medical diagnostics and healthcare informatics.

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

Optimized Bi-GRU Models for Enhanced Heart Disease Prediction Through Data Mining Techniques

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • B. Manjurathi,
  • A. Madhumathi,
  • P. Karthikeyan,
  • P. Saravanakumar,
  • S. Kannadhasan

摘要

The Anticipation of heart diseases has garnered considerable scholarly interest due to its potential to preserve life through the facilitation of prompt identification and intervention. The aim of this effort is to create an optimal Bidirectional Gated Recurrent Unit (Bi-GRU) model that accurately predicts cardiac illness using data mining approaches. We provide the Grey Wolf Optimization (GWO) method, which effectively adjusts the Bi-GRU model’s hyper parameters to escalate performance. The dataset’s objective variable indicates the presence or absence of cardiac disease. Age, sex, type of chest pain (CP), Cholesterol (chol), fasting blood sugar (fbs), resting blood pressure (trestbps), and other clinical characteristics are also included. The Bi-GRU architecture is utilised to identify temporal correlations in the patient data after feature extraction and pre-processing, while GWO optimises factors including batch size, learning rate, and hidden unit count. The optimized Bi-GRU model outperforms conventional models in terms of prediction and reliability, according to experimental data. Ultimately, by showing how bio-inspired optimization and advanced DL architectures may be applied to escalate the prediction of cardiac illness, our approach enhances the domains of medical diagnostics and healthcare informatics.