Skip to main content
SpringerLink
Log in

Prediction of preoperative microvascular invasion by dynamic radiomic analysis based on contrast-enhanced computed tomography

  • Interventional Radiology
  • Published:
Abdominal Radiology Aims and scope Submit manuscript

Abstract

Purpose

Microvascular invasion (MVI) is a common complication of hepatocellular carcinoma (HCC) surgery, which is an important predictor of reduced surgical prognosis. This study aimed to develop a fully automated diagnostic model to predict pre-surgical MVI based on four-phase dynamic CT images.

Methods

A total of 140 patients with HCC from two centers were retrospectively included (training set, n = 98; testing set, n = 42). All CT phases were aligned to the portal venous phase, and were then used to train a deep-learning model for liver tumor segmentation. Radiomics features were extracted from the tumor areas of original CT phases and pairwise subtraction images, as well as peritumoral features. Lastly, linear discriminant analysis (LDA) models were trained based on clinical features, radiomics features, and hybrid features, respectively. Models were evaluated by area under curve (AUC), accuracy, sensitivity, specificity, positive and negative predictive values (PPV and NPV).

Results

Overall, 86 and 54 patients with MVI− (age, 55.92 ± 9.62 years; 68 men) and MVI+ (age, 53.59 ± 11.47 years; 43 men) were included. Average dice coefficients of liver tumor segmentation were 0.89 and 0.82 in training and testing sets, respectively. The model based on radiomics (AUC = 0.865, 95% CI: 0.725–0.951) showed slightly better performance than that based on clinical features (AUC = 0.841, 95% CI: 0.696–0.936). The classification model based on hybrid features achieved better performance in both training (AUC = 0.955, 95% CI: 0.893–0.987) and testing sets (AUC = 0.913, 95% CI: 0.785–0.978), compared with models based on clinical and radiomics features (p-value < 0.05). Moreover, the hybrid model also provided the best accuracy (0.857), sensitivity (0.875), and NPV (0.917).

Conclusion

The classification model based on multimodal intra- and peri-tumoral radiomics features can well predict HCC patients with MVI.

Graphical Abstract

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6

Similar content being viewed by others

References

  1. Sperandio RC, Pestana RC, Miyamura BV, Kaseb AO (2022) Hepatocellular carcinoma immunotherapy. Annu Rev Med 73:267–278. https://doi.org/10.1146/annurev-med-042220-021121

    Article  CAS  PubMed  Google Scholar 

  2. Romain D, Amaia L (2022) The liver cancer immune microenvironment: Therapeutic implications for hepatocellular carcinoma. HEPATOLOGY. https://doi.org/10.1002/hep.32740

    Article  Google Scholar 

  3. Lima HA, Moazzam Z, Endo Y, et al (2023) TBS-Based preoperative score to predict non-transplantable recurrence and identify candidates for upfront resection versus transplantation for hepatocellular carcinoma. Ann Surg Oncol. https://doi.org/10.1245/s10434-023-13273-1

    Article  PubMed  Google Scholar 

  4. Wei Y, Pei W, Qin Y, et al (2021) Preoperative MR imaging for predicting early recurrence of solitary hepatocellular carcinoma without microvascular invasion. Eur J Radiol 138:109663. https://doi.org/10.1016/j.ejrad.2021.109663

    Article  PubMed  Google Scholar 

  5. Imura S, Teraoku H, Yoshikawa M, et al (2018) Potential predictive factors for microvascular invasion in hepatocellular carcinoma classified within the Milan criteria. Int J Clin Oncol 23:98–103. https://doi.org/10.1007/s10147-017-1189-8

    Article  PubMed  Google Scholar 

  6. Yang C, Liu X, Ling W-W, et al (2020) Primary isolated hepatic tuberculosis mimicking small hepatocellular carcinoma. Medicine. https://doi.org/10.1097/MD.0000000000022580

    Article  PubMed  PubMed Central  Google Scholar 

  7. Xie D, Ren Z, Zhou J, et al (2020) 2019 Chinese clinical guidelines for the management of hepatocellular carcinoma: updates and insights. Hepatobiliary Surg Nutr v.9(4). https://doi.org/10.21037/hbsn-20-480

  8. Song L, Li J, Luo Y (2021) The importance of a nonsmooth tumor margin and incomplete tumor capsule in predicting HCC microvascular invasion on preoperative imaging examination: a systematic review and meta-analysis. Clin Imaging 76:77–82. https://doi.org/10.1016/j.clinimag.2020.11.057

    Article  PubMed  Google Scholar 

  9. Zhang H-M, Wen D-G, Chen J, et al (2023) A diagnostic test of three-dimensional magnetic resonance elastography imaging for preoperative prediction of microvascular invasion in patients with T1 stage clear cell renal carcinoma. Transl Androl Urol 12. https://doi.org/10.21037/tau-23-94

  10. Peng J, Zhang J, Zhang Q-F, et al (2018) A radiomics nomogram for preoperative prediction of microvascular invasion risk in hepatitis B virus-related hepatocellular carcinoma. Diagn Interv Radiol. https://doi.org/10.5152/dir.2018.17467

    Article  PubMed  PubMed Central  Google Scholar 

  11. Lambin P, Leijenaar RT, Deist TM, et al (2017) Radiomics: the bridge between medical imaging and personalized medicine. Nat Rev Clin Oncol 14:749–762. https://doi.org/10.1038/nrclinonc.2017.141

    Article  PubMed  Google Scholar 

  12. Jha AK, Mithun S, Purandare NC, Kumar R, et al (2022) Radiomics: a quantitative imaging biomarker in precision oncology. Nucl Med Commun 43:483–493. https://doi.org/10.1097/MNM.0000000000001543

    Article  PubMed  Google Scholar 

  13. Jiang Y-Q, Cao S-E, Cao S, et al (2021) Preoperative identification of microvascular invasion in hepatocellular carcinoma by XGBoost and deep learning. J Cancer Res Clin Oncol 147:821–833. https://doi.org/10.1007/s00432-020-03366-9

    Article  PubMed  Google Scholar 

  14. Jiang C, Zhao L, Xin B, et al (2022) 18F-FDG PET/CT radiomic analysis for classifying and predicting microvascular invasion in hepatocellular carcinoma and intrahepatic cholangiocarcinoma. Quant Imaging Med Surg 12:4135–4150. https://doi.org/10.21037/qims-21-1167

  15. Park S, Kim JH, Kim J, et al (2023) Development of a deep learning-based auto-segmentation algorithm for hepatocellular carcinoma (HCC) and application to predict microvascular invasion of HCC using CT texture analysis: preliminary results. Acta Radiol 64:907–917. https://doi.org/10.1177/02841851221100318

    Article  PubMed  Google Scholar 

  16. Yao W, Yang S, Ge Y, et al (2022) Computed tomography radiomics-based prediction of microvascular invasion in hepatocellular carcinoma. Front Med 9. https://doi.org/10.3389/fmed.2022.819670

  17. Azam M, Khan K, Ahmad M, et al (2021) Multimodal medical image registration and fusion for quality enhancement. Comput Mater Contin 68:821–840. https://doi.org/10.32604/cmc.2021.016131

  18. Albers J, Svetlove A, Alves J, et al (2021) Elastic transformation of histological slices allows precise co-registration with microCT data sets for a refined virtual histology approach. Sci Rep 11:10846. https://doi.org/10.1038/s41598-021-89841-w

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Rietzel E, Pan T, Chen GTY (2005) Four-dimensional computed tomography: Image formation and clinical protocol. Med Phys 32:874–889. https://doi.org/10.1118/1.1869852

    Article  PubMed  Google Scholar 

  20. Decaux N, Conze P-H, Ropars J, et al (2023) Semi-automatic muscle segmentation in MR images using deep registration-based label propagation. Pattern Recognit 140:109529. https://doi.org/10.1016/j.patcog.2023.109529

    Article  PubMed  Google Scholar 

  21. Li L, Zhao X, Lu W, Tan S (2020) Deep learning for variational multimodality tumor segmentation in PET/CT. Neurocomputing 392:277–295. https://doi.org/10.1016/j.neucom.2018.10.099

    Article  PubMed  Google Scholar 

  22. Zhang Y, Peng C, Peng L, et al (2021) Multi-phase liver tumor segmentation with spatial aggregation and uncertain region inpainting, Medical Image Computing and Computer Assisted Intervention. https://doi.org/10.1007/978-3-030-87193-2_7

    Article  PubMed  PubMed Central  Google Scholar 

  23. Aoki T, Kamiya T, Lu H, et al (2021) CT temporal subtraction: techniques and clinical applications. Quant Imaging Med Surg 11:2214–2223. https://doi.org/10.21037/qims-20-1367

  24. Kim DH, Choi SH, Byun JH, et al (2019) Arterial subtraction images of gadoxetate-enhanced MRI improve diagnosis of early-stage hepatocellular carcinoma. J Hepatol 71:534–542. https://doi.org/10.1016/j.jhep.2019.05.005

    Article  CAS  PubMed  Google Scholar 

  25. Lee J, Kim KW, Kim SY, et al (2015) Automatic detection method of hepatocellular carcinomas using the non-rigid registration method of multi-phase liver CT images. J X-Ray Sci Technol 23:275–288. https://doi.org/10.3233/XST-150487

    Article  CAS  Google Scholar 

  26. Zhang T, Pandey G, Xu L, et al (2020) The value of TTPVI in prediction of microvascular invasion in hepatocellular carcinoma. Cancer Manag Res 12:4097. https://doi.org/10.2147/CMAR.S245475

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  27. Peng J, Zhang J, Zhang Q, et al (2018) A radiomics nomogram for preoperative prediction of microvascular invasion risk in hepatitis B virus-related hepatocellular carcinoma. Diagn Interv Radiol 24:121. https://doi.org/10.5152/dir.2018.17467

    Article  PubMed  PubMed Central  Google Scholar 

  28. Jiang Y, Cao S, Cao S, et al (2021) Preoperative identifcation of microvascular invasion in hepatocellular carcinoma by XGBoost and deep learning. Journal of Cancer Research and Clinical Oncology 147:821–833. https://doi.org/10.1007/s00432-020-03366-9

    Article  PubMed  Google Scholar 

  29. Wu X, Dong D, Zhang L, et al (2021) Exploring the predictive value of additional peritumoral regions based on deep learning and radiomics: a multicenter study. Med Phys 48:2374–2385. https://doi.org/10.1002/mp.14767

    Article  PubMed  Google Scholar 

  30. Chong H-H, Yang L, Sheng R-F, et al (2021) Multi-scale and multi-parametric radiomics of gadoxetate disodium–enhanced MRI predicts microvascular invasion and outcome in patients with solitary hepatocellular carcinoma≤ 5 cm. Eur Radiol 31:4824–4838. https://doi.org/10.1007/s00330-020-07601-2

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  31. Chong H, Gong Y, Pan X, et al (2021) Peritumoral dilation radiomics of gadoxetate disodium-enhanced MRI excellently predicts early recurrence of hepatocellular carcinoma without macrovascular invasion after hepatectomy. J Hepatocell Carcinoma 545–563. https://doi.org/10.2147/JHC.S309570

  32. Zhang W, Yang R, Liang F, et al (2021) Prediction of microvascular invasion in hepatocellular carcinoma with a multi-Disciplinary team-Like radiomics fusion model on dynamic contrast-Enhanced computed tomography. Front Oncol 11:660629. https://doi.org/10.3389/fonc.2021.660629

    Article  PubMed  PubMed Central  Google Scholar 

  33. Shipe ME, Deppen SA, Farjah F, Grogan EL (2019) Developing prediction models for clinical use using logistic regression: an overview. J Thorac Dis 11:S574. https://doi.org/10.21037/jtd.2019.01.25

  34. Azad TD, Ehresman J, Ahmed AK, et al (2021) Fostering reproducibility and generalizability in machine learning for clinical prediction modeling in spine surgery. Spine J 21:1610–1616. https://doi.org/10.1016/j.spinee.2020.10.006

    Article  PubMed  Google Scholar 

  35. Steyerberg EW, Steyerberg EW (2019) Overfitting and optimism in prediction models. Clin Predict Models Pract Approach. https://doi.org/10.1007/978-3-030-16399-0_5

    Article  Google Scholar 

  36. Guo Y, Mokany K, Ong C, et al (2023) Plant species richness prediction from DESIS hyperspectral data: A comparison study on feature extraction procedures and regression models. ISPRS J Photogramm Remote Sens 196:120–133. https://doi.org/10.1016/j.isprsjprs.2022.12.028

    Article  Google Scholar 

Download references

Funding

This work was supported by National Natural Science Foundation of China (Nos. 62171230, 62101365, 92159301, 91959207, 62301263, 62301265, 62302228, 82302291, 82302352).

Author information

Author notes

  1. Zhenghao Zhou and Tianyi Xia have contributed equally to this work as co-first authors.

Authors and Affiliations

  1. School of Artificial Intelligence, Institute for AI in Medicine, Nanjing University of Information Science and Technology, Nanjing, 210044, China

    Zhenghao Zhou, Teng Zhang, Yunzhi Huang, Kai Xuan, Zhuo Wan & Jun Xu

  2. Jiangsu Key Laboratory of Molecular and Functional Imaging, Department of Radiology, School of Medicine, Zhongda Hospital, Southeast University, 87 Ding Jia Qiao Road, Nanjing, 210009, China

    Tianyi Xia, Jiarui Zhong & Shenghong Ju

  3. Cerebrovascular Disease Treatment Center, Nanjing Brain Hospital Affiliated to Brain Hospital of Nanjing Medical University, Nanjing, 210029, China

    Mingyang Du

  4. Information School, University of Washington, Seattle, WA, 98195, USA

    Geyang Xu

Authors
  1. Zhenghao Zhou
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Tianyi Xia
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Teng Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Mingyang Du
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Jiarui Zhong
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Yunzhi Huang
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Kai Xuan
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Geyang Xu
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Zhuo Wan
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Shenghong Ju
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Jun Xu
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding authors

Correspondence to Shenghong Ju or Jun Xu.

Ethics declarations

Conflict of interest

The authors of this manuscript declare no relationships with each other and any companies, whose products or services may be related to the subject matter of the article.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary file1 (DOCX 175 kb)

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zhou, Z., Xia, T., Zhang, T. et al. Prediction of preoperative microvascular invasion by dynamic radiomic analysis based on contrast-enhanced computed tomography. Abdom Radiol 49, 611–624 (2024). https://doi.org/10.1007/s00261-023-04102-w

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00261-023-04102-w

Keywords

Search

Navigation