108384
论文已发表
注册即可获取德孚的最新动态
IF 收录期刊
- 3.4 Breast Cancer (Dove Med Press)
- 3.2 Clin Epidemiol
- 2.6 Cancer Manag Res
- 2.9 Infect Drug Resist
- 3.7 Clin Interv Aging
- 5.1 Drug Des Dev Ther
- 3.1 Int J Chronic Obstr
- 6.6 Int J Nanomed
- 2.6 Int J Women's Health
- 2.9 Neuropsych Dis Treat
- 2.8 OncoTargets Ther
- 2.0 Patient Prefer Adher
- 2.2 Ther Clin Risk Manag
- 2.5 J Pain Res
- 3.0 Diabet Metab Synd Ob
- 3.2 Psychol Res Behav Ma
- 3.4 Nat Sci Sleep
- 1.8 Pharmgenomics Pers Med
- 2.0 Risk Manag Healthc Policy
- 4.1 J Inflamm Res
- 2.0 Int J Gen Med
- 3.4 J Hepatocell Carcinoma
- 3.0 J Asthma Allergy
- 2.2 Clin Cosmet Investig Dermatol
- 2.4 J Multidiscip Healthc
已发表论文
基于 CT 的 2.5D 深度学习 - 多示例学习用于预测肝细胞癌的早期复发并关联复发相关病理指标
Authors Cen Y, Nong H, Du D, Wu Y, Chen J, Pan Z, Huang Y, Ding K, Huang D
Received 4 June 2025
Accepted for publication 29 August 2025
Published 17 September 2025 Volume 2025:12 Pages 2095—2108
DOI https://doi.org/10.2147/JHC.S541402
Checked for plagiarism Yes
Review by Single anonymous peer review
Peer reviewer comments 3
Editor who approved publication: Dr Ali Hosni
Yongyi Cen,1,2,* Haiyang Nong,1,2,* Dehui Du,3,* Yingning Wu,1,2 Jianpeng Chen,1,2 Zhaolin Pan,4 Yin Huang,5 Ke Ding,6 Deyou Huang1,2
1Department of Radiology, Affiliated Hospital of Youjiang Medical University for Nationalities, Baise, Guangxi Zhuang Autonomous Region, 533000, People’s Republic of China; 2Guangxi Clinical Medical Research Center for Hepatobiliary Diseases, Affiliated Hospital of Youiiang Medical University for Nationalities, Baise, Guangxi Zhuang Autonomous Region, 533000, People’s Republic of China; 3Department of Radiology, The Affiliated Wuming Hospital of Guangxi Medical University, Nanning, Guangxi Zhuang Autonomous Region, 530100, People’s Republic of China; 4Department of Hepatobiliary Surgery, Affiliated Hospital of Youjiang Medical University for Nationalities, Baise, Guangxi Zhuang Autonomous Region, 533000, People’s Republic of China; 5Pathology Department, Affiliated Hospital of Youjiang University of Nationalities, Baise, Guangxi Zhuang Autonomous Region, 533000, People’s Republic of China; 6Department of Radiology, The Third Affiliated Hospital of Guangxi Medical University, Nanning, Guangxi Zhuang Autonomous Region, 530031, People’s Republic of China
*These authors contributed equally to this work
Correspondence: Deyou Huang, Department of Radiology, Affiliated Hospital of Youjiang Medical University for Nationalities, Baise, Guangxi Zhuang Autonomous Region, 533000, People’s Republic of China, Email FZXYH2012@126.com Ke Ding, Department of Radiology, The Third Affiliated Hospital of Guangxi Medical University, Nanning, Guangxi Zhuang Autonomous Region, 530031, People’s Republic of China, Email 272480365@qq.com
Purpose: This study aims to evaluate the advantages of the 2.5D deep learning-multi-instance learning (2.5D DL-MIL) model, based on CT arterial phase images, in predicting early recurrence (ER) of hepatocellular carcinoma (HCC) and examining the biological significance of MIL features.
Patients and Methods: A total of 191 HCC patients were retrospectively included and categorized into ER (n=79) and non-early recurrence (NER, n=112) groups based on postoperative follow-up results. The patients were randomly divided to the training set (n=133) and validation set (n=58) in a 7:3 ratio. The predictive capabilities of the 2.5D DL-MIL model, Radiomics model, and Clinical model for ER of HCC were constructed and compared using CT arterial phase and clinical data. SHAP analysis was used to evaluate the contribution of MIL features in the model, and further analysis was conducted on the correlation between MIL features and microvascular invasion (MVI), Ki-67 expression, and pathological grading.
Results: The area under the curve (AUC) for the 2.5D DL-MIL model in the validation set was 0.840, surpassing that of the Radiomics model (AUC = 0.678, P = 0.047) and the Clinical model (AUC = 0.598, P = 0.009). Decision curve analyses indicated superior clinical utility for the 2.5D DL-MIL model. SHAP analysis revealed that bag-of-words features (eg, BoW_02 and BoW_09) were key contributors to the 2.5D DL-MIL model. Correlation analysis demonstrated that BoW_01, BoW_02, BoW_09, and BoW_1 were significantly correlated with MVI grade and Ki-67 expression (P < 0.05).
Conclusion: The 2.5D DL-MIL model demonstrates significant value in predicting ER of HCC, with its MIL features exhibiting strong associations with tumor invasiveness and proliferative activity.
Keywords: hepatocellular carcinoma, early recurrence, CT, deep learning, multi-instance learning
1Department of Radiology, Affiliated Hospital of Youjiang Medical University for Nationalities, Baise, Guangxi Zhuang Autonomous Region, 533000, People’s Republic of China; 2Guangxi Clinical Medical Research Center for Hepatobiliary Diseases, Affiliated Hospital of Youiiang Medical University for Nationalities, Baise, Guangxi Zhuang Autonomous Region, 533000, People’s Republic of China; 3Department of Radiology, The Affiliated Wuming Hospital of Guangxi Medical University, Nanning, Guangxi Zhuang Autonomous Region, 530100, People’s Republic of China; 4Department of Hepatobiliary Surgery, Affiliated Hospital of Youjiang Medical University for Nationalities, Baise, Guangxi Zhuang Autonomous Region, 533000, People’s Republic of China; 5Pathology Department, Affiliated Hospital of Youjiang University of Nationalities, Baise, Guangxi Zhuang Autonomous Region, 533000, People’s Republic of China; 6Department of Radiology, The Third Affiliated Hospital of Guangxi Medical University, Nanning, Guangxi Zhuang Autonomous Region, 530031, People’s Republic of China
*These authors contributed equally to this work
Correspondence: Deyou Huang, Department of Radiology, Affiliated Hospital of Youjiang Medical University for Nationalities, Baise, Guangxi Zhuang Autonomous Region, 533000, People’s Republic of China, Email FZXYH2012@126.com Ke Ding, Department of Radiology, The Third Affiliated Hospital of Guangxi Medical University, Nanning, Guangxi Zhuang Autonomous Region, 530031, People’s Republic of China, Email 272480365@qq.com
Purpose: This study aims to evaluate the advantages of the 2.5D deep learning-multi-instance learning (2.5D DL-MIL) model, based on CT arterial phase images, in predicting early recurrence (ER) of hepatocellular carcinoma (HCC) and examining the biological significance of MIL features.
Patients and Methods: A total of 191 HCC patients were retrospectively included and categorized into ER (n=79) and non-early recurrence (NER, n=112) groups based on postoperative follow-up results. The patients were randomly divided to the training set (n=133) and validation set (n=58) in a 7:3 ratio. The predictive capabilities of the 2.5D DL-MIL model, Radiomics model, and Clinical model for ER of HCC were constructed and compared using CT arterial phase and clinical data. SHAP analysis was used to evaluate the contribution of MIL features in the model, and further analysis was conducted on the correlation between MIL features and microvascular invasion (MVI), Ki-67 expression, and pathological grading.
Results: The area under the curve (AUC) for the 2.5D DL-MIL model in the validation set was 0.840, surpassing that of the Radiomics model (AUC = 0.678, P = 0.047) and the Clinical model (AUC = 0.598, P = 0.009). Decision curve analyses indicated superior clinical utility for the 2.5D DL-MIL model. SHAP analysis revealed that bag-of-words features (eg, BoW_02 and BoW_09) were key contributors to the 2.5D DL-MIL model. Correlation analysis demonstrated that BoW_01, BoW_02, BoW_09, and BoW_1 were significantly correlated with MVI grade and Ki-67 expression (P < 0.05).
Conclusion: The 2.5D DL-MIL model demonstrates significant value in predicting ER of HCC, with its MIL features exhibiting strong associations with tumor invasiveness and proliferative activity.
Keywords: hepatocellular carcinoma, early recurrence, CT, deep learning, multi-instance learning