Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2021 Aug;46(8):3835-3844.
doi: 10.1007/s00261-021-03027-6. Epub 2021 Mar 17.

A radiomic-based model of different contrast-enhanced CT phase for differentiate intrahepatic cholangiocarcinoma from inflammatory mass with hepatolithiasis

Beihui Xue #  1 , Sunjie Wu #  1 , Mingyue Zhang  1 , Junjie Hong  1 , Bole Liu  1 , Nina Xu  1 , Qiqiang Zeng  2 , Kun Tang  1 , Xiangwu Zheng  3
Affiliations

A radiomic-based model of different contrast-enhanced CT phase for differentiate intrahepatic cholangiocarcinoma from inflammatory mass with hepatolithiasis

Beihui Xue et al. Abdom Radiol (NY). 2021 Aug.

Abstract

Background: Intrahepatic cholangiocarcinoma (ICC) is hard to distinguish from inflammatory mass (IM) complicated with hepatolithiasis in clinical practice preoperatively. This study looked to develop and confirm the radiomics models to make a distinction between ICC with hepatolithiasis from IM and to compare the results of different contrast-enhanced computed tomography (CT) phase.

Methods: The models were developed in a training cohort of 110 patients from January 2005 to June 2020. Radiomics features were extracted from both arterial phase and portal venous phase contrast-enhanced computed tomography (CT) scans. The radiomics scores based on radiomics features, were built by logistic regression after using the least absolute shrinkage and selection operator (LASSO) method. The rad-scores of two contrast -enhanced CT phases and clinical features were incorporated into a novel model. The performance of the models were determined by theirs discrimination, calibration, and clinical usefulness. The models were externally validated in 35 consecutive patients.

Results: The radiomics signature comprised two features in arterial phase (training cohort, AUC = 0.809, sensitivity 0.700, specificity 0.848, and accuracy 0.774;validation cohort, AUC = 0.790, sensitivity 0.714, specificity 0.800, and accuracy 0.757) and three related features in portal venous phase (training cohort, AUC = 0.801, sensitivity 0.800, specificity 0.717, and accuracy 0.759; validation cohort, AUC = 0.830, sensitivity 0.700, specificity 0.750, and accuracy 0.775) showed significant association with ICC in both cohorts (P < 0.05).We also developed a model only based on clinical variables (training cohort, AUC = 0.778, sensitivity 0.567, specificity 0.891, and accuracy 0.729; validation cohort, AUC = 0.788, sensitivity 0.571, specificity 0.950, and accuracy 0.761). The radiomics-based model contained rad-score of two phases and two clinical factors (CEA and CA19-9) showed the best performance (training cohort, AUC = 0.864, sensitivity 0.867, specificity 0.804, and accuracy 0.836; validation cohort, AUC = 0.843, sensitivity 0.643, specificity 0.980, and accuracy 0.821).

Conclusions: Our radiomics-based models provided a diagnostic tool for differentiate intrahepatic cholangiocarcinoma (ICC) from inflammatory mass (IM) with hepatolithiasis both in arterial phase and portal venous phase. To go a step further, the diagnostic accuracy will improved by a clinico-radiologic model.

Keywords: Hepatolithiasis; Inflammatory mass(IM); Intrahepatic cholangiocarcinoma (ICC); Radiomics.

PubMed Disclaimer

References

    1. S. Rizvi, S.A. Khan, C.L. Hallemeier, R.K. Kelley, G.J. Gores, Cholangiocarcinoma - evolving concepts and therapeutic strategies, Nat Rev Clin Oncol 15(2) (2018) 95-111. - DOI
    1. R. Wood, D.H. Brewster, L.A. Fraser, H. Brown, P.C. Hayes, O.J. Garden, Do increases in mortality from intrahepatic cholangiocarcinoma reflect a genuine increase in risk? Insights from cancer registry data in Scotland, Eur J Cancer 39(14) (2003) 2087-92. - DOI
    1. B. Sripa, C. Pairojkul, Cholangiocarcinoma: lessons from Thailand, Curr Opin Gastroenterol 24(3) (2008) 349-56. - DOI
    1. Y. Shaib, H.B. El-Serag, The epidemiology of cholangiocarcinoma, Semin Liver Dis 24(2) (2004) 115-25. - DOI
    1. I. Joo, J.M. Lee, J.H. Yoon, Imaging Diagnosis of Intrahepatic and Perihilar Cholangiocarcinoma: Recent Advances and Challenges, Radiology 288(1) (2018) 7-23. - DOI

LinkOut - more resources