Scholarlink

Effective decision-making is essential in healthcare, and the increasing use of Artificial Intelligence (AI) is driving significant advancements in the medical field [13]. Explainable Artificial Intelligence (XAI) has emerged as a key factor in ensuring transparency and interpretability in AI-driven healthcare applications. This study focuses on the role of XAI in enhancing understanding and trust in areas such as medical image classification, disease prediction, and diagnostic tasks. In particular, the application of XAI techniques is crucial for making these complex AI processes more transparent and interpretable. This paper reviews recent advancements in XAI techniques, including Local Interpretable Model-Agnostic Explanations (LIME), SHapley Additive exPlanations (SHAP), and Gradient-weighted Class Activation Mapping (Grad-CAM), Layer-wise relevance propagation (LRP), Contextual Importance and Utility (CIU), Continuous Fuzzy Cognitive Map Classifier (CFCMC) and NeuroXAI techniques highlighting their impact on improving the interpretability of healthcare AI systems and fostering more reliable clinical decision-making.

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 Review of Explainable AI Techniques for Enhancing Interpretability in Healthcare

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • K. Hemalatha,
  • G. Rama Saketh,
  • Jeevesh Pranav Ravikannan,
  • G. Jayanth

摘要

Effective decision-making is essential in healthcare, and the increasing use of Artificial Intelligence (AI) is driving significant advancements in the medical field [13]. Explainable Artificial Intelligence (XAI) has emerged as a key factor in ensuring transparency and interpretability in AI-driven healthcare applications. This study focuses on the role of XAI in enhancing understanding and trust in areas such as medical image classification, disease prediction, and diagnostic tasks. In particular, the application of XAI techniques is crucial for making these complex AI processes more transparent and interpretable. This paper reviews recent advancements in XAI techniques, including Local Interpretable Model-Agnostic Explanations (LIME), SHapley Additive exPlanations (SHAP), and Gradient-weighted Class Activation Mapping (Grad-CAM), Layer-wise relevance propagation (LRP), Contextual Importance and Utility (CIU), Continuous Fuzzy Cognitive Map Classifier (CFCMC) and NeuroXAI techniques highlighting their impact on improving the interpretability of healthcare AI systems and fostering more reliable clinical decision-making.