Scholarlink

<p>Breast cancer (BC) remains one of the foremost reasons of mortality worldwide. Early detection of BC can significantly support timely diagnosis and helpful in treatment. This study presents a fresh improved version of Chernobyl disaster optimizer (CDO) known as ICDO method for classifying BC disease. CDO version has been merged with Dimension Learning Hunting (DLH) search strategy to reduce shortcomings of CDO method for instance diversity, poor local avoidance, slower and premature convergence, weak exploration ability and failure in trapping the goal respectively. The ICDO method benefits from DLH strategy to enhance the exploitation and exploration behaviour of particles in search area. Secondly, this work also presents the classification model based on a merged neural network, ICDO, along with exchange knowledge for classifying BC datasets. To evaluate the strength of ICDO method, 29-CEC’ 2017 benchmarks and two different datasets such as MIAS and CBIS-DDSM have been used in this study. Tabulated results reveal that ICDO model is superior to others in proving the best optimal solutions for CEC functions and classification results for datasets.</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

Performance Evaluation of Improved Version of Chernobyl Disaster Optimizer for Classifying Breast Cancer

  • 期刊论文

  • 发表日期：

    2026年4月28日

查看全文
  • Bhawna Utreja,
  • Reecha Sharma,
  • Amit Wason

摘要

Breast cancer (BC) remains one of the foremost reasons of mortality worldwide. Early detection of BC can significantly support timely diagnosis and helpful in treatment. This study presents a fresh improved version of Chernobyl disaster optimizer (CDO) known as ICDO method for classifying BC disease. CDO version has been merged with Dimension Learning Hunting (DLH) search strategy to reduce shortcomings of CDO method for instance diversity, poor local avoidance, slower and premature convergence, weak exploration ability and failure in trapping the goal respectively. The ICDO method benefits from DLH strategy to enhance the exploitation and exploration behaviour of particles in search area. Secondly, this work also presents the classification model based on a merged neural network, ICDO, along with exchange knowledge for classifying BC datasets. To evaluate the strength of ICDO method, 29-CEC’ 2017 benchmarks and two different datasets such as MIAS and CBIS-DDSM have been used in this study. Tabulated results reveal that ICDO model is superior to others in proving the best optimal solutions for CEC functions and classification results for datasets.