Scholarlink

<p>Breast cancer recurrence remains a significant challenge in oncology, impacting patient prognosis and treatment strategies. Early and accurate prediction of recurrence can enhance clinical decision-making and improve patient outcomes. In this study, we propose a novel Optimized Prediction Algorithm (OPA) that integrates attention mechanisms, active learning, and eXplainable Artificial Intelligence (XAI) to predict breast cancer recurrence. The proposed OPA algorithm was trained and validated on a comprehensive dataset of breast cancer patients, achieving 94.5% accuracy, 92.7% sensitivity, and 95.8% specificity, outperforming existing models. Furthermore, the incorporation of XAI techniques ensures model interpretability by providing transparent insights into key predictive factors, thereby increasing clinical trust and applicability. The OPA algorithm holds great potential for assisting oncologists in identifying high-risk patients and enabling timely intervention for improved patient care.</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

An explainable AI-powered algorithm for predicting breast cancer recurrence using attention mechanisms and active learning

  • 期刊论文

  • 发表日期：

    2026年5月26日

查看全文
  • Fatma M. Talaat,
  • Mohamed Abdelkhalek,
  • Saleh Saleh Elbalka,
  • Ahmed Abdallah

摘要

Breast cancer recurrence remains a significant challenge in oncology, impacting patient prognosis and treatment strategies. Early and accurate prediction of recurrence can enhance clinical decision-making and improve patient outcomes. In this study, we propose a novel Optimized Prediction Algorithm (OPA) that integrates attention mechanisms, active learning, and eXplainable Artificial Intelligence (XAI) to predict breast cancer recurrence. The proposed OPA algorithm was trained and validated on a comprehensive dataset of breast cancer patients, achieving 94.5% accuracy, 92.7% sensitivity, and 95.8% specificity, outperforming existing models. Furthermore, the incorporation of XAI techniques ensures model interpretability by providing transparent insights into key predictive factors, thereby increasing clinical trust and applicability. The OPA algorithm holds great potential for assisting oncologists in identifying high-risk patients and enabling timely intervention for improved patient care.