Deep learning in differentiating the colorectal cancer combined with hepatic enhancing nodules: liver metastases vs hemangiomas
TotalSegmentator can perform automatic segmentation of CRLMs and HMs, but demonstrates poorer segmentation consistency for subcentimeter lesions. This DL model assists radiologists in distinguishing 10-30 mm CRLMs from HMs in CRC patients. Subcentimeter CRLMs and HMs can require further MRI scanning.
Authors
- Jialiang Ren
- ShengLin Li
- Fengyu Zhou (ORCID: https://orcid.org/0000-0001-5140-7036)
- Hao Wang (ORCID: https://orcid.org/0000-0001-6881-9977)
- Guojin Zhang (ORCID: https://orcid.org/0000-0002-4312-5212)
- Xinmei Yang (ORCID: https://orcid.org/0000-0003-3537-4582)
- Yuebo Wang
- Junlin Zhou
- Yaqiong Ma
- Long Yuan
- Yuan Xu
- Yu Guo
- Zhimei Jiao
- Yufeng Li
- Ting Lu
Institutions
- University of Electronic Science and Technology of China (CN)
- Chengdu Medical College (CN)
- Hexi University (CN)
- Lanzhou University Second Hospital (CN)
- Artificial Intelligence in Medicine (Canada) (CA)
- Lanzhou University (CN)
Publication Details
- Journal
- Insights into Imaging
- Published
- 2026-01-26
- DOI
- https://doi.org/10.1186/s13244-025-02192-2
- Primary Topic
- Advanced X-ray and CT Imaging
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Hexi University
- Lanzhou University
- Chengdu Medical College
- Jiangsu Development and Reform Commission
- University of Electronic Science and Technology of China
- Natural Science Foundation of Sichuan Province