A CNN‐transformer‐based hybrid U‐shape model with long‐range relay for esophagus 3D CT image gross tumor volume segmentation
We propose a CNN-Transformer hybrid deep learning network to improve the segmentation effect of esophageal tumors. We utilize the local and global information between shallower and deeper layers to prevent early information loss and enhance the cross-layer communication. To validate our model, we collect a dataset composed of 1665 CT images of esophageal tumors from Sichuan Tumor Hospital. The results show that our model outperforms the state-of-the-art models. It is of great significance to improve the accuracy and clinical application of esophageal tumor segmentation.
Authors
- Pengfei Jiao (ORCID: https://orcid.org/0000-0003-1049-1002)
- Yixiu Liu (ORCID: https://orcid.org/0000-0002-4630-3774)
- Chenggang Yan (ORCID: https://orcid.org/0000-0001-5957-3207)
- Shuai Wang (ORCID: https://orcid.org/0000-0002-1570-6570)
- Yunxiang Li (ORCID: https://orcid.org/0000-0003-3244-2228)
- Jianxiang Zhao
- Qifeng Wang (ORCID: https://orcid.org/0000-0002-7880-0476)
- Songli Yu
Institutions
- University of Electronic Science and Technology of China (CN)
- Shandong University (CN)
- Southwestern Medical Center (US)
- Sichuan Cancer Hospital (CN)
- Suzhou Research Institute (CN)
- Southwestern Medical Center (US)
- Hangzhou Dianzi University (CN)
- The University of Texas Southwestern Medical Center (US)
Publication Details
- Journal
- Medical Physics
- Published
- 2025-04-14
- DOI
- https://doi.org/10.1002/mp.17818
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- National Natural Science Foundation of China