Cross-institutional evaluation of deep learning and radiomics models in predicting microvascular invasion in hepatocellular carcinoma: validity, robustness, and ultrasound modality efficacy comparison
Cross-institutional validation indicated that DL demonstrated better robustness than radiomics for preoperative MVI prediction in patients with HCC, representing a promising solution to non-standardized ultrasound examination procedures.
Authors
- Qiang Guo (ORCID: https://orcid.org/0000-0002-2326-2244)
- Guangwen Cheng (ORCID: https://orcid.org/0000-0002-8831-9213)
- Tong Zhang (ORCID: https://orcid.org/0000-0001-7602-3856)
- Weibin Zhang
- Yuli Zhu (ORCID: https://orcid.org/0009-0008-1584-8772)
- Hong Ding (ORCID: https://orcid.org/0000-0003-3753-3241)
- Qi Zhang (ORCID: https://orcid.org/0000-0002-8433-7625)
- Meng Wang (ORCID: https://orcid.org/0009-0003-3191-2272)
Institutions
- Shanghai University (CN)
- Fudan University (CN)
- Zhongshan Hospital (CN)
- Huashan Hospital (CN)
Publication Details
- Journal
- Cancer Imaging
- Published
- 2024-10-22
- DOI
- https://doi.org/10.1186/s40644-024-00790-9
- Citations
- 23
- Primary Topic
- Hepatocellular Carcinoma Treatment and Prognosis
- Type
- article
- Field-Weighted Citation Impact
- 8.50
Funders
- Shanghai Science and Technology Development Foundation
- National Natural Science Foundation of China