Scholarlink

As one of the primary causes of cancer-related death globally, colorectal cancer (CRC) highlights the importance of precise and effective screening techniques. With an emphasis on preprocessing methods and classifier performance assessment, this work explores deep learning for CRC screening using colonoscopy images. The preprocessing pipeline consists of scaling, normalization, mask removal, and principal component analysis (PCA) to improve image quality and feature extraction. As the primary classifier, a Convolutional Neural Network (CNN) is used, and its performance is evaluated against cutting-edge deep learning models such as Xception, GoogleNet, and ResNet. The usefulness of the suggested method was demonstrated by the trials, which were carried out using Python and produced an impressive accuracy of 96.53%. These results imply that deep learning-based techniques have great potential to increase the precision and dependability of colorectal cancer screening.

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

Deep Learning for Colorectal Cancer Screening Using Colonoscopy Images

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Mandar K. Mokashi,
  • Ganesh Shelke,
  • Manali M. Shah,
  • Mahesh S. Shinde,
  • Mahesh Bhandari,
  • Dattatray G. Takale,
  • Parikshit N. Mahalle,
  • Bipin Sule

摘要

As one of the primary causes of cancer-related death globally, colorectal cancer (CRC) highlights the importance of precise and effective screening techniques. With an emphasis on preprocessing methods and classifier performance assessment, this work explores deep learning for CRC screening using colonoscopy images. The preprocessing pipeline consists of scaling, normalization, mask removal, and principal component analysis (PCA) to improve image quality and feature extraction. As the primary classifier, a Convolutional Neural Network (CNN) is used, and its performance is evaluated against cutting-edge deep learning models such as Xception, GoogleNet, and ResNet. The usefulness of the suggested method was demonstrated by the trials, which were carried out using Python and produced an impressive accuracy of 96.53%. These results imply that deep learning-based techniques have great potential to increase the precision and dependability of colorectal cancer screening.