Scholarlink

<p>Breast cancer can be cured, especially if discovered early, to enhance the survival prospects of the patient. Past research has employed several deep learning techniques, but their detection abilities are often limited in detecting breast cancer. To find solutions to these limitations, an innovative solution known as the Efficient Cascade One-Dimensional Shuffle Navigated Network with Influencer Buddy Optimization (ECOD-SNNet-IBO) is proposed. It is tested with mammogram images from the INbreast, BreakHis, and MIAS (Mammographic Image Analysis Society) databases. To enhance image quality by reducing noise and artifacts, it preprocesses images with a Gradient Guided Filter (GGF). Efficient Cascade Convolution (ECC) can be used to segment breast cancer regions with accuracy. Classification is addressed with the Shuffle Navigated Neural Network (SNNNet) after feature retrieval with the One-Dimensional Discrete Transform (ODDT). The performance of the model is enhanced by Influencer Buddy Optimization (IBO) and making it more robust and efficient. With an accuracy of 99.9% and a recall rate of 99.8%, the ECOD-SNNet-IBO method achieves remarkable results in the INbreast, BreakHis, and MIAS datasets. These results outperform those of existing methods, indicating the joint possibility of achieving high diagnostic accuracy of lesion type, particularly cancer, and improving the diagnostic precision of early breast cancer (BC), particularly in mammography.</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

Breast Cancer Detection Using Efficient Cascade One-Dimensional Shuffle Navigated Network with Influencer Buddy Optimization

  • 期刊论文

  • 发表日期：

    2025年10月19日

查看全文
  • Anwar Ahamed Shaikh,
  • Ajaypradeep Natarajsivam,
  • P. Shanmuga Prabha,
  • Elangovan Muniyandy

摘要

Breast cancer can be cured, especially if discovered early, to enhance the survival prospects of the patient. Past research has employed several deep learning techniques, but their detection abilities are often limited in detecting breast cancer. To find solutions to these limitations, an innovative solution known as the Efficient Cascade One-Dimensional Shuffle Navigated Network with Influencer Buddy Optimization (ECOD-SNNet-IBO) is proposed. It is tested with mammogram images from the INbreast, BreakHis, and MIAS (Mammographic Image Analysis Society) databases. To enhance image quality by reducing noise and artifacts, it preprocesses images with a Gradient Guided Filter (GGF). Efficient Cascade Convolution (ECC) can be used to segment breast cancer regions with accuracy. Classification is addressed with the Shuffle Navigated Neural Network (SNNNet) after feature retrieval with the One-Dimensional Discrete Transform (ODDT). The performance of the model is enhanced by Influencer Buddy Optimization (IBO) and making it more robust and efficient. With an accuracy of 99.9% and a recall rate of 99.8%, the ECOD-SNNet-IBO method achieves remarkable results in the INbreast, BreakHis, and MIAS datasets. These results outperform those of existing methods, indicating the joint possibility of achieving high diagnostic accuracy of lesion type, particularly cancer, and improving the diagnostic precision of early breast cancer (BC), particularly in mammography.