Preoperative prediction for early recurrence of hepatocellular carcinoma using machine learning-based radiomics

Bing Mao; Yajun Ren; Xuan Yu; Xinliang Liang; Xiangming Ding

doi:10.3389/fonc.2024.1346124

Preoperative prediction for early recurrence of hepatocellular carcinoma using machine learning-based radiomics

Front Oncol. 2024 Mar 15:14:1346124. doi: 10.3389/fonc.2024.1346124. eCollection 2024.

Authors

Bing Mao^#¹, Yajun Ren^#², Xuan Yu³, Xinliang Liang¹, Xiangming Ding⁴

Affiliations

¹ Henan Provincial People's Hospital, Zhengzhou University People's Hospital; Henan University People's Hospital, Zhengzhou, Henan, China.
² Department of Gastroenterology, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, China.
³ Department of Medical Imaging, Henan Provincial People's Hospital, People's Hospital of Zhengzhou University, Zhengzhou, China.
⁴ Department of Gastroenterology, Henan Provincial People's Hospital, People's Hospital of Zhengzhou University, Zhengzhou, Henan, China.

^# Contributed equally.

Abstract

Objective: To develop a contrast-enhanced computed tomography (CECT) based radiomics model using machine learning method and assess its ability of preoperative prediction for the early recurrence of hepatocellular carcinoma (HCC).

Methods: A total of 297 patients confirmed with HCC were assigned to the training dataset and test dataset based on the 8:2 ratio, and the follow-up period of the patients was from May 2012 to July 2017. The lesion sites were manually segmented using ITK-SNAP, and the pyradiomics platform was applied to extract radiomic features. We established the machine learning model to predict the early recurrence of HCC. The accuracy, AUC, standard deviation, specificity, and sensitivity were applied to evaluate the model performance.

Results: 1,688 features were extracted from the arterial phase and venous phase images, respectively. When arterial phase and venous phase images were employed correlated with clinical factors to train a prediction model, it achieved the best performance (AUC with 95% CI 0.8300(0.7560-0.9040), sensitivity 89.45%, specificity 79.07%, accuracy 82.67%, p value 0.0064).

Conclusion: The CECT-based radiomics may be helpful to non-invasively reveal the potential connection between CECT images and early recurrence of HCC. The combination of radiomics and clinical factors could boost model performance.

Keywords: early recurrence; hepatocellular carcinoma; machine learning; multidetector computed tomography; radiomics.

Grants and funding

The author(s) declare financial support was received for the research, authorship, and/or publication of this article. The study was funded by Medical Science & Technology Research Program of Henan Province (Grant NO. RKX202202002), Henan Province Medical Science and Technology Research Project Joint Construction Project (LHGJ20190256), and National Natural Science Foundation of China (Grant No. 82103617).