Skip to main content
SpringerLink
Log in

Distinguishing common renal cell carcinomas from benign renal tumors based on machine learning: comparing various CT imaging phases, slices, tumor sizes, and ROI segmentation strategies

  • Imaging Informatics and Artificial Intelligence
  • Published:
European Radiology Aims and scope Submit manuscript

Abstract

Objectives

To determine whether a CT-based machine learning (ML) can differentiate benign renal tumors from renal cell carcinomas (RCCs) and improve radiologists’ diagnostic performance, and evaluate the impact of variable CT imaging phases, slices, tumor sizes, and region of interest (ROI) segmentation strategies.

Methods

Patients with pathologically proven RCCs and benign renal tumors from our institution between 2008 and 2020 were included as the training dataset for ML model development and internal validation (including 418 RCCs and 78 benign tumors), and patients from two independent institutions and a public database (TCIA) were included as the external dataset for individual testing (including 262 RCCs and 47 benign tumors). Features were extracted from three-phase CT images. CatBoost was used for feature selection and ML model establishment. The area under the receiver operating characteristic curve (AUC) was used to assess the performance of the ML model.

Results

The ML model based on 3D images performed better than that based on 2D images, with the highest AUC of 0.81 and accuracy (ACC) of 0.86. All three radiologists achieved better performance by referring to the classifier’s decision, with accuracies increasing from 0.82 to 0.87, 0.82 to 0.88, and 0.76 to 0.87. The ML model achieved higher negative predictive values (NPV, 0.82–0.99), and the radiologists achieved higher positive predictive values (PPV, 0.91–0.95).

Conclusions

A ML classifier based on whole-tumor three-phase CT images can be a useful and promising tool for differentiating RCCs from benign renal tumors. The ML model also perfectly complements radiologist interpretations.

Key Points

• A machine learning classifier based on CT images could be a reliable way to differentiate RCCs from benign renal tumors.

• The machine learning model perfectly complemented the radiologists’ interpretations.

• Subtle variances in ROI delineation had little effect on the performance of the ML classifier.

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

Similar content being viewed by others

Abbreviations

ACC:

Accuracy

AML:

Angiomyolipoma

ASL:

Arterial spin labeling

CECT:

Contrast-enhanced computed tomography

CMP:

Corticomedullary phase

DWI:

Diffusion-weighted imaging

GLCM:

Gray-level cooccurrence matrix

GLDM:

Gray-level dependence matrix

GLRLM:

Gray-level run-length matrix

GLSZM:

Gray-level size-zone matrix

ML:

Machine learning

NGTDM:

Neighborhood gray-tone difference matrix

NP:

Nephrographic phase

PCP:

Precontrast phase

RCC:

Renal cell carcinoma

ROI:

Region of interest

References

  1. Vendrami CL, Villavicencio CP, Dejulio TJ et al (2017) Differentiation of solid renal tumors with multiparametric MR imaging. Radiographics 37(7):2026–2042. https://doi.org/10.1148/rg.2017170039

  2. Volpe A, Panzarella T, Rendon RA, Haider MA, Kondylis FI, Jewett MA (2004) The natural history of incidentally detected small renal masses. Cancer 100(4):738–745. https://doi.org/10.1002/cncr.20025

    Article  PubMed  Google Scholar 

  3. Gill IS, Aron M, Gervais DA, Jewett MAS (2010) Clinical practice. Small renal mass. N Engl J Med 362(7):624–634. https://doi.org/10.1056/NEJMcp0910041

    Article  CAS  PubMed  Google Scholar 

  4. Lane BR, Babineau D, Kattan MW et al (2007) A preoperative prognostic nomogram for solid enhancing renal tumors 7 cm or less amenable to partial nephrectomy. J Urol 178(2):429–434. https://doi.org/10.1016/j.juro.2007.03.106

    Article  PubMed  Google Scholar 

  5. Lane BR, Samplaski MK, Herts BR, Zhou M, Novick AC, Campbell SC (2008) Renal mass biopsy--a renaissance? J Urol 179(1):20–27. https://doi.org/10.1016/j.juro.2007.08.124

    Article  PubMed  Google Scholar 

  6. de Leon AD, Kapur P, Pedrosa I (2019) Radiomics in kidney cancer: MR imaging. Magn Reson Imaging Clin N Am 27(1):1–13. https://doi.org/10.1016/j.mric.2018.08.005

    Article  PubMed  Google Scholar 

  7. Erdim C, Yardimci AH, Bektas CT et al (2020) Prediction of benign and malignant solid renal masses: machine learning-based CT texture analysis. Acad Radiol 27(10):1422–1429. https://doi.org/10.1016/j.acra.2019.12.015

    Article  PubMed  Google Scholar 

  8. Gillies RJ, Kinahan PE, Hricak H (2016) Radiomics: images are more than pictures, they are data. Radiology 278(2):563–577. https://doi.org/10.1148/radiol.2015151169

    Article  PubMed  Google Scholar 

  9. Davnall F, Yip CSP, Ljungqvist G et al (2012) Assessment of tumor heterogeneity: an emerging imaging tool for clinical practice? Insights Imaging 3(6):573–589. https://doi.org/10.1007/s13244-012-0196-6

    Article  PubMed  PubMed Central  Google Scholar 

  10. Suarez-Ibarrola R, Basulto-Martinez M, Heinze A, Gratzke C, Miernik A (2020) Radiomics applications in renal tumor assessment: a comprehensive review of the literature. Cancers (Basel) 12(6):1–25. https://doi.org/10.3390/cancers12061387

    Article  CAS  Google Scholar 

  11. Tabibu S, Vinod PK, Jawahar CV (2019) Pan-renal cell carcinoma classification and survival prediction from histopathology images using deep learning. Sci Rep 9(1):1–9. https://doi.org/10.1038/s41598-019-46718-3

    Article  CAS  Google Scholar 

  12. Sun XY, Feng QX, Xu X et al (2020) Radiologic-radiomic machine learning models for differentiation of benign and malignant solid renal masses: comparison with expert-level radiologists. AJR Am J Roentgenol 214(1):W44–W54. https://doi.org/10.2214/AJR.19.21617

  13. Cui E, Li Z, Ma C et al (2020) Predicting the ISUP grade of clear cell renal cell carcinoma with multiparametric MR and multiphase CT radiomics. Eur Radiol 30(5):2912–2921. https://doi.org/10.1007/s00330-019-06601-1

    Article  PubMed  Google Scholar 

  14. Yushkevich PA, Piven J, Hazlett HC et al (2006) User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability. Neuroimage 31(3):1116–1128. https://doi.org/10.1016/j.neuroimage.2006.01.015

    Article  PubMed  Google Scholar 

  15. Cui EM, Lin F, Li Q et al (2019) Differentiation of renal angiomyolipoma without visible fat from renal cell carcinoma by machine learning based on whole-tumor computed tomography texture features. Acta Radiol 60(11):1543–1552. https://doi.org/10.1177/0284185119830282

  16. Chen F, Gulati M, Hwang D et al (2017) Voxel-based whole-lesion enhancement parameters: a study of its clinical value in differentiating clear cell renal cell carcinoma from renal oncocytoma. Abdom Radiol (NY) 42(2):552–560. https://doi.org/10.1007/s00261-016-0891-8

  17. Mazzei FG, Mazzei MA, Cioffi Squitieri N et al (2014) CT perfusion in the characterisation of renal lesions: an added value to multiphasic CT. Biomed Res Int 2014:135013. https://doi.org/10.1155/2014/135013

    Article  PubMed  PubMed Central  Google Scholar 

  18. Haralick RM, Shanmugam K, Dinstein I (1973) Textural features for image classification. Stud Media Commun SMC 3(6):610–621. 10.1109/TSMC.1973.4309314.

  19. Varghese BA, Chen F, Hwang DH (2018) Differentiation of predominantly solid enhancing lipid-poor renal cell masses by use of contrast-enhanced CT: evaluating the role of texture in tumor subtyping. AJR Am J Roentgenol 211(6):W288–W296. https://doi.org/10.2214/AJR.18.19551

    Article  PubMed  Google Scholar 

  20. van Griethuysen JJM, Fedorov A, Parmar C et al (2017) Computational radiomics system to decode the radiographic phenotype. Cancer Res 77(21):e104–e107. https://doi.org/10.1158/0008-5472.CAN-17-0339

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  21. Prokhorenkova L, Gusev G, Vorobev A, Dorogush AV, Gulin A (2018) Catboost: unbiased boosting with categorical features. Adv Neural Inf Process Syst 2018-Decem:6638–6648. https://doi.org/10.48550/arXiv.1706.09516

  22. Dorogush AV, Ershov V, Gulin A (2018) CatBoost: gradient boosting with categorical features support. 1–7. https://doi.org/10.48550/arXiv.1810.11363

  23. Dyer R, DiSantis DJ, McClennan BL (2008) Simplified imaging approach for evaluation of the solid renal mass in adults. Radiology 247(2):331–343. https://doi.org/10.1148/radiol.2472061846

    Article  PubMed  Google Scholar 

  24. Leveridge MJ, Bostrom PJ, Koulouris G, Finelli A, Lawrentschuk N (2010) Imaging renal cell carcinoma with ultrasonography, CT and MRI. Nat Rev Urol 7(6):311–325. https://doi.org/10.1038/nrurol.2010.63

    Article  PubMed  Google Scholar 

  25. Kutikov A, Smaldone MC, Egleston BL et al (2011) Anatomic features of enhancing renal masses predict malignant and high-grade pathology: a preoperative nomogram using the RENAL Nephrometry score. Eur Urol 60(2):241–248. https://doi.org/10.1016/j.eururo.2011.03.029

    Article  PubMed  PubMed Central  Google Scholar 

  26. Hunter JD (2007) Matplotlib: a 2D graphics environment. Comput Sci Eng 9(3):90–95. https://doi.org/10.1109/MCSE.2007.55

    Article  Google Scholar 

  27. Kunapuli G, Varghese BA, Ganapathy P et al (2018) A decision-support tool for renal mass classification. J Digit Imaging 31(6):929–939. https://doi.org/10.1007/s10278-018-0100-0

    Article  PubMed  PubMed Central  Google Scholar 

  28. Xi IL, Zhao Y, Wang R et al (2020) Deep learning to distinguish benign from malignant renal lesions based on routine MR imaging. Clin Cancer Res 26(8):1944–1952. https://doi.org/10.1158/1078-0432.CCR-19-0374

    Article  PubMed  Google Scholar 

  29. Tanaka T, Huang Y, Marukawa Y et al (2020) Differentiation of small (≤ 4 cm) renal masses on multiphase contrast-enhanced CT by deep learning. AJR Am J Roentgenol 214(3):605–612. https://doi.org/10.2214/AJR.19.22074

Download references

Acknowledgements

I would like to express my gratitude to Professor Wansheng Long, my supervisor, for his constant encouragement, guidance, and enlightening instructions, which contributed to the completion of my paper. I would also like to acknowledge my indebtedness to Enming Cui and Fan Lin, without whose valuable resources and support this paper could not have appeared in its final form. I am also grateful to all the other teachers and classmates who have given me generous support and helpful advice in the course of shaping my paper. Last but not least, my thanks go to my beloved families, for their loving considerations and great confidence in me all through these years.

Funding

This study has received funding from Guangdong Basic and Applied Basic Research Foundation (Grant Number: 2021A1515220080) and the Opening Research Fund of Guangzhou Key Laboratory of Molecular and Functional Imaging for Clinical Translation (Grant Number: 201905010003).

Author information

Author notes

  1. Tao Zhou and Jian Guan contributed equally to this study.

Authors and Affiliations

  1. Department of Radiology, Jiangmen Central Hospital, Guangdong Medical University, Zunyi Medical University, 23 Beijie Haibang Street, Jiangmen, 529030, People’s Republic of China

    Tao Zhou, Bao Feng, Huimin Xue, Jin Cui, Qionglian Kuang, Enming Cui & Wansheng Long

  2. Department of Radiology, The First Affiliated Hospital of Sun Yat-Sen University, 58 Zhongshan Second Road, Guangzhou, 510000, People’s Republic of China

    Tao Zhou, Jian Guan & Qionglian Kuang

  3. Guangzhou Key Laboratory of Molecular and Functional Imaging for Clinical Translation, Guangzhou, People’s Republic of China

    Bao Feng, Huimin Xue, Enming Cui & Wansheng Long

  4. School of Electronic Information and Automation, Guilin University of Aerospace Technology, 2 Jinji Road, Guilin, 541000, People’s Republic of China

    Yehang Chen & Kuncai Xu

  5. Department of Radiology, The First Affiliated Hospital of Shenzhen University, Health Science Center, Shenzhen Second People’s Hospital, 3002 SunGangXi Road, Shenzhen, 518035, People’s Republic of China

    Fan Lin

Authors
  1. Tao Zhou
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Jian Guan
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Bao Feng
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Huimin Xue
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Jin Cui
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Qionglian Kuang
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Yehang Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Kuncai Xu
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Fan Lin
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Enming Cui
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Wansheng Long
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding authors

Correspondence to Fan Lin, Enming Cui or Wansheng Long.

Ethics declarations

Guarantor

The scientific guarantor of this publication is Wansheng Long.

Conflict of interest

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

Statistics and biometry

No complex statistical methods were necessary for this paper.

Informed consent

Written informed consent was waived by the Institutional Review Board.

Ethical approval

Institutional Review Board approval was obtained.

Methodology

• retrospective

• diagnostic or prognostic study

• multicenter study

Additional information

Publisher’s note

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

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, T., Guan, J., Feng, B. et al. Distinguishing common renal cell carcinomas from benign renal tumors based on machine learning: comparing various CT imaging phases, slices, tumor sizes, and ROI segmentation strategies. Eur Radiol 33, 4323–4332 (2023). https://doi.org/10.1007/s00330-022-09384-0

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00330-022-09384-0

Keywords

Search

Navigation