Scholarlink

Background <p>Circulating tumor cells (CTCs) in peripheral blood are crucial for prognosis, treatment response, disease monitoring, and personalized therapy. However, identifying CTCs remains challenging due to their scarcity and heterogeneity, even with advanced deep learning models.</p> Methods <p>This study introduces an innovative hybrid framework combining a dual-branch network with traditional image processing techniques and automated CTC identification. By incorporating image and fluorescence attributes, the framework enhances feature representation robustness. Performance was evaluated using accuracy, precision, and recall metrics and comparisons with pathologists’ manual counting.</p> Results <p>The framework achieved 97.05% accuracy in distinguishing CTCs from non-CTCs, with performance closely matching pathologists’ manual counting in survival prediction. The dual-branch network improved efficiency by leveraging segmentation algorithms, surpassing conventional methods. Clinical trials confirmed its practicality for direct clinical use.</p> Conclusions <p>The proposed framework enhances CTC identification accuracy and efficiency, demonstrating strong clinical applicability. Its output results can be directly utilized for prognosis without manual intervention, offering significant potential for personalized therapy.</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

A dual-branch deep learning network for circulating tumor cells classification

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年9月24日

查看全文
  • Chao Han,
  • Jiaquan Lin,
  • Yanfang Liang,
  • Cong Li,
  • Danni Wang,
  • Gonghua Huang,
  • Ruoxi Hong,
  • Jincheng Zeng

摘要

Background

Circulating tumor cells (CTCs) in peripheral blood are crucial for prognosis, treatment response, disease monitoring, and personalized therapy. However, identifying CTCs remains challenging due to their scarcity and heterogeneity, even with advanced deep learning models.

Methods

This study introduces an innovative hybrid framework combining a dual-branch network with traditional image processing techniques and automated CTC identification. By incorporating image and fluorescence attributes, the framework enhances feature representation robustness. Performance was evaluated using accuracy, precision, and recall metrics and comparisons with pathologists’ manual counting.

Results

The framework achieved 97.05% accuracy in distinguishing CTCs from non-CTCs, with performance closely matching pathologists’ manual counting in survival prediction. The dual-branch network improved efficiency by leveraging segmentation algorithms, surpassing conventional methods. Clinical trials confirmed its practicality for direct clinical use.

Conclusions

The proposed framework enhances CTC identification accuracy and efficiency, demonstrating strong clinical applicability. Its output results can be directly utilized for prognosis without manual intervention, offering significant potential for personalized therapy.