Scholarlink

<p>Accurate and efficient classification of histopathological images is essential for improving colon cancer diagnosis and treatment planning. This paper introduces Dilated Residual SqueezeNet (DRCS-Net), an enhanced deep learning framework for colon cancer diagnosis using histopathological images. By integrating dilated convolutions and residual connections within the SqueezeNet architecture, DRCS-Net effectively captures multi-scale features and contextual information. Experiments on e LC25000 dataset demonstrate 99.80% accuracy, outperforming standard models such as SqueezeNet and Residual SqueezeNet. The results highlight the impact of architectural innovations on colon cancer classification accuracy. To foster further research, the source code and dataset are made openly available.</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

Improved colon cancer diagnosis with a dilated residual SqueezeNet architecture: a deep learning approach

  • 期刊论文

  • 发表日期：

    2025年5月27日

查看全文
  • R. Kishor,
  • R. S. Vinod Kumar,
  • A. R. Bushara,
  • D. Shahi

摘要

Accurate and efficient classification of histopathological images is essential for improving colon cancer diagnosis and treatment planning. This paper introduces Dilated Residual SqueezeNet (DRCS-Net), an enhanced deep learning framework for colon cancer diagnosis using histopathological images. By integrating dilated convolutions and residual connections within the SqueezeNet architecture, DRCS-Net effectively captures multi-scale features and contextual information. Experiments on e LC25000 dataset demonstrate 99.80% accuracy, outperforming standard models such as SqueezeNet and Residual SqueezeNet. The results highlight the impact of architectural innovations on colon cancer classification accuracy. To foster further research, the source code and dataset are made openly available.