Scholarlink

<p>The heart is a vital organ of the human body and requires proper care to prevent life-threatening conditions. Accurate prediction of heart disease is essential, as delayed diagnosis continues to impact patients adversely. Despite advancements in the field, existing prediction algorithms often lack the precision and efficiency needed for reliable outcomes. A key contribution of this research is the integration of Explainable Artificial Intelligence (XAI) techniques, specifically SHAP (SHapley Additive exPlanations), to enhance model transparency. SHAP analysis provided fine-grained insights into feature importance, ensuring that model predictions align with clinical reasoning and enabling healthcare professionals to interpret results effectively. By combining high accuracy with interpretability, this study demonstrates the potential of AI-driven predictive systems in supporting early diagnosis and risk assessment of heart disease. The findings emphasize the importance of explainable, data-driven approaches in enhancing clinical decision-making, reducing misdiagnosis rates, and improving patient outcomes. Using a random forest classifier with hyperparameter tuning, the ML approach achieved 99% accuracy, with precision, recall, and F1-scores of 0.99. The confusion matrix revealed 159 true negatives and 146 true positives. Meanwhile, the DL approach employed artificial neural networks (ANNs) enhanced with an attention mechanism, achieving 98% accuracy, with precision, recall, and F1-scores of 0.98, and a confusion matrix of 157 true negatives and 146 true positives. The findings underscore the potential of both approaches in improving heart disease prediction while maintaining model transparency.</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

The role of explainable AI in building trust and acceptance of AI-driven heart disease

  • 期刊论文

  • 发表日期：

    2025年9月15日

查看全文
  • Fatma M. Talaat,
  • Wesam F. Aly,
  • Rana Mohamed El-Balka,
  • Mohamed Shehata,
  • Samah Adel Gamel

摘要

The heart is a vital organ of the human body and requires proper care to prevent life-threatening conditions. Accurate prediction of heart disease is essential, as delayed diagnosis continues to impact patients adversely. Despite advancements in the field, existing prediction algorithms often lack the precision and efficiency needed for reliable outcomes. A key contribution of this research is the integration of Explainable Artificial Intelligence (XAI) techniques, specifically SHAP (SHapley Additive exPlanations), to enhance model transparency. SHAP analysis provided fine-grained insights into feature importance, ensuring that model predictions align with clinical reasoning and enabling healthcare professionals to interpret results effectively. By combining high accuracy with interpretability, this study demonstrates the potential of AI-driven predictive systems in supporting early diagnosis and risk assessment of heart disease. The findings emphasize the importance of explainable, data-driven approaches in enhancing clinical decision-making, reducing misdiagnosis rates, and improving patient outcomes. Using a random forest classifier with hyperparameter tuning, the ML approach achieved 99% accuracy, with precision, recall, and F1-scores of 0.99. The confusion matrix revealed 159 true negatives and 146 true positives. Meanwhile, the DL approach employed artificial neural networks (ANNs) enhanced with an attention mechanism, achieving 98% accuracy, with precision, recall, and F1-scores of 0.98, and a confusion matrix of 157 true negatives and 146 true positives. The findings underscore the potential of both approaches in improving heart disease prediction while maintaining model transparency.