Scholarlink

Lung cancer remains a significant global health challenge, with early detection being crucial for improving patient outcomes. This study explores the use of transfer learning in enhancing the diagnostic accuracy of lung cancer classification from CT images. Specifically, it compares the effectiveness of two transfer learning pipelines utilizing pre-trained models, VGG19 and ResNet152, as feature extractors combined with Logistic Regression (LR) for classification. The results demonstrate that the VGG16 + LR pipeline outperforms the ResNet152 pipeline, achieving superior classification accuracy for both validation and testing datasets at 97%, respectively. This finding underscores the potential of transfer learning in medical imaging, offering a promising approach to improving early detection and treatment strategies for lung cancer.

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

Leveraging Transfer Learning as Feature Extractors for Lung Cancer Classification: Insights from VGG19 and ResNet152 Pipelines

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Darren Soong Kai Xuan,
  • Anwar P.P. Abdul Majeed,
  • Rabiu Muazu Musa,
  • Yang Luo,
  • Saad Aslam,
  • Samuel-Soma M. Ajibade,
  • Mehran Behjati,
  • Muhammed Basheer Jasser,
  • Muhammad Amirul Abdullah

摘要

Lung cancer remains a significant global health challenge, with early detection being crucial for improving patient outcomes. This study explores the use of transfer learning in enhancing the diagnostic accuracy of lung cancer classification from CT images. Specifically, it compares the effectiveness of two transfer learning pipelines utilizing pre-trained models, VGG19 and ResNet152, as feature extractors combined with Logistic Regression (LR) for classification. The results demonstrate that the VGG16 + LR pipeline outperforms the ResNet152 pipeline, achieving superior classification accuracy for both validation and testing datasets at 97%, respectively. This finding underscores the potential of transfer learning in medical imaging, offering a promising approach to improving early detection and treatment strategies for lung cancer.