Top

Published in:

01-01-2021 | Kidney Cancer | Kidneys, Ureters, Bladder, Retroperitoneum

Stratification of cystic renal masses into benign and potentially malignant: applying machine learning to the bosniak classification

Authors: Nityanand Miskin, Lei Qin, Shanna A. Matalon, Sree H. Tirumani, Francesco Alessandrino, Stuart G. Silverman, Atul B. Shinagare

Published in: Abdominal Radiology | Issue 1/2021

Abstract

Purpose

To create a CT texture-based machine learning algorithm that distinguishes benign from potentially malignant cystic renal masses as defined by the Bosniak Classification version 2019.

Methods

In this IRB-approved, HIPAA-compliant study, 4,454 adult patients underwent renal mass protocol CT or CT urography from January 2011 to June 2018. Of these, 257 cystic renal masses were included in the final study cohort. Each mass was independently classified using Bosniak version 2019 by three radiologists, resulting in 185 benign (Bosniak I or II) and 72 potentially malignant (Bosniak IIF, III or IV) masses. Six texture features: mean, standard deviation, mean of positive pixels, entropy, skewness, kurtosis were extracted using commercial software TexRAD (Feedback PLC, Cambridge, UK). Random forest (RF), logistic regression (LR), and support vector machine (SVM) machine learning algorithms were implemented to classify cystic renal masses into the two groups and tested with tenfold cross validations.

Results

Higher mean, standard deviation, mean of positive pixels, entropy, skewness were statistically associated with the potentially malignant group (P ≤ 0.0015 each). Sensitivity, specificity, positive predictive value, negative predictive value, and area under curve of RF model was 0.67, 0.91, 0.75, 0.88, 0.88; of LR model was 0.63, 0.93, 0.78, 0.86, 0.90, and of SVM model was 0.56, 0.91, 0.71, 0.84, 0.89, respectively.

Conclusion

Three CT texture-based machine learning algorithms demonstrated high discriminatory capability in distinguishing benign from potentially malignant cystic renal masses as defined by the Bosniak Classification version 2019. If validated, CT texture-based machine learning algorithms may help reduce interreader variability when applying the Bosniak classification.

Terada N, Arai Y, Kinukawa N, Yoshimura K, Terai A (2004) Risk factors for renal cysts. BJU Int 93:1300–1302. https://doi.org/10.1111/j.1464-410X.2004.04844.xCrossRefPubMed

Carrim ZI, Murchison JT (2003) The prevalence of simple renal and hepatic cysts detected by spiral computed tomography. Clin Radiol 58:626–629CrossRefPubMed

Suher M, Koc E, Bayrak G (2006) Simple renal cyst prevalence in internal medicine department and concomitant diseases. Ren Fail 28:149–152CrossRefPubMed

Tada S, Yamagishi J, Kobayashi H, Hata Y, Kobari T (1983) The incidence of simple renal cyst by computed tomography. Clin Radiol 34:437–439CrossRefPubMed

Siegel RL, Miller KD, Jemal A (2019) Cancer statistics, 2019. CA Cancer J Clin 69:7–34. https://doi.org/10.3322/caac.21551CrossRefPubMed

Krajewski KM, Pedrosa I (2018) Imaging Advances in the Management of Kidney Cancer. J Clin Oncol Off J Am Soc Clin Oncol JCO2018791236. https://doi.org/10.1200/JCO.2018.79.1236

Davies L, Petitti DB, Woo M, Lin JS (2018) Defining, Estimating, and Communicating Overdiagnosis in Cancer Screening. Ann Intern Med 169:824. https://doi.org/10.7326/L18-0517CrossRefPubMed

Esserman LJ, Thompson IM, Reid B (2013) Overdiagnosis and overtreatment in cancer: an opportunity for improvement. JAMA 310:797–798. https://doi.org/10.1001/jama.2013.108415CrossRefPubMed

Schoots IG, Zaccai K, Hunink MG, Verhagen PCMS (2017) Bosniak Classification for Complex Renal Cysts Reevaluated: A Systematic Review. J Urol 198:12–21. https://doi.org/10.1016/j.juro.2016.09.160CrossRefPubMed

10.

Go AS, Chertow GM, Fan D, McCulloch CE, Hsu C (2004) Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. N Engl J Med 351:1296–1305. https://doi.org/10.1056/NEJMoa041031CrossRefPubMed

11.

Sun M, Thuret R, Abdollah F, Lughezzani G, Schmitges J, Tian Z, Shariat SF, Montorsi F, Patard J-J, Perrotte P, Karakiewicz PI (2011) Age-adjusted incidence, mortality, and survival rates of stage-specific renal cell carcinoma in North America: a trend analysis. Eur Urol 59:135–141. https://doi.org/10.1016/j.eururo.2010.10.029CrossRefPubMed

12.

Sun M, Trinh Q-D, Bianchi M, Hansen J, Hanna N, Abdollah F, Shariat SF, Briganti A, Montorsi F, Perrotte P, Karakiewicz PI (2012) A non-cancer-related survival benefit is associated with partial nephrectomy. Eur Urol 61:725–731. https://doi.org/10.1016/j.eururo.2011.11.047CrossRefPubMed

13.

Tan H-J, Norton EC, Ye Z, Hafez KS, Gore JL, Miller DC (2012) Long-term survival following partial vs radical nephrectomy among older patients with early-stage kidney cancer. JAMA 307:1629–1635. https://doi.org/10.1001/jama.2012.475CrossRefPubMed

14.

Van Poppel H, Da Pozzo L, Albrecht W, Matveev V, Bono A, Borkowski A, Colombel M, Klotz L, Skinner E, Keane T, Marreaud S, Collette S, Sylvester R (2011) A prospective, randomised EORTC intergroup phase 3 study comparing the oncologic outcome of elective nephron-sparing surgery and radical nephrectomy for low-stage renal cell carcinoma. Eur Urol 59:543–552. https://doi.org/10.1016/j.eururo.2010.12.013CrossRefPubMed

15.

Silverman SG, Pedrosa I, Ellis JH, Hindman NM, Schieda N, Smith AD, Remer EM, Shinagare AB, Curci NE, Raman SS, Wells SA, Kaffenberger SD, Wang ZJ, Chandarana H, Davenport MS (2019) Bosniak Classification of Cystic Renal Masses, Version 2019: An Update Proposal and Needs Assessment. Radiology 292:475–488. https://doi.org/10.1148/radiol.2019182646CrossRefPubMed

16.

Hu EM, Zhang A, Silverman SG, Pedrosa I, Wang ZJ, Smith AD, Chandarana H, Doshi A, Shinagare AB, Remer EM, Kaffenberger SD, Miller DC, Davenport MS (2018) Multi-institutional analysis of CT and MRI reports evaluating indeterminate renal masses: comparison to a national survey investigating desired report elements. Abdom Radiol N Y 43:3493–3502. https://doi.org/10.1007/s00261-018-1609-xCrossRef

17.

Hindman NM, Hecht EM, Bosniak MA (2014) Follow-up for Bosniak category 2F cystic renal lesions. Radiology 272:757–766. https://doi.org/10.1148/radiol.14122908CrossRefPubMed

18.

Hindman NM (2016) Cystic renal masses. Abdom Radiol N Y 41:1020–1034. https://doi.org/10.1007/s00261-016-0761-4CrossRef

19.

Silverman SG, Gan YU, Mortele KJ, Tuncali K, Cibas ES (2006) Renal masses in the adult patient: the role of percutaneous biopsy. Radiology 240:6–22. https://doi.org/10.1148/radiol.2401050061CrossRefPubMed

20.

Weibl P, Klatte T, Kollarik B, Waldert M, Schüller G, Geryk B, Remzi M (2011) Interpersonal variability and present diagnostic dilemmas in Bosniak classification system. Scand J Urol Nephrol 45:239–244. https://doi.org/10.3109/00365599.2011.562233CrossRefPubMed

21.

Siegel CL, McFarland EG, Brink JA, Fisher AJ, Humphrey P, Heiken JP (1997) CT of cystic renal masses: analysis of diagnostic performance and interobserver variation. AJR Am J Roentgenol 169:813–818. https://doi.org/10.2214/ajr.169.3.9275902CrossRefPubMed

22.

Siegel CL, Fisher AJ, Bennett HF (1999) Interobserver variability in determining enhancement of renal masses on helical CT. AJR Am J Roentgenol 172:1207–1212. https://doi.org/10.2214/ajr.172.5.10227490CrossRefPubMed

23.

Benjaminov O, Atri M, O’Malley M, Lobo K, Tomlinson G (2006) Enhancing component on CT to predict malignancy in cystic renal masses and interobserver agreement of different CT features. AJR Am J Roentgenol 186:665–672. https://doi.org/10.2214/AJR.04.0372CrossRefPubMed

24.

Kim DY, Kim JK, Min G-E, Ahn H-J (1987) Cho K-S (2010) Malignant renal cysts: diagnostic performance and strong predictors at MDCT. Acta Radiol Stockh Swed 51:590–598. https://doi.org/10.3109/02841851003641826CrossRef

25.

El-Mokadem I, Budak M, Pillai S, Lang S, Doull R, Goodman C, Nabi G (2014) Progression, interobserver agreement, and malignancy rate in complex renal cysts ( ≥ Bosniak category IIF). Urol Oncol 32:24.e21–27. https://doi.org/10.1016/j.urolonc.2012.08.018CrossRef

26.

Graumann O, Osther SS, Karstoft J, Hørlyck A (1987) Osther PJS (2015) Bosniak classification system: inter-observer and intra-observer agreement among experienced uroradiologists. Acta Radiol Stockh Swed 56:374–383. https://doi.org/10.1177/0284185114529562CrossRef

27.

Summers RM (2017) Texture analysis in radiology: Does the emperor have no clothes? Abdom Radiol N Y 42:342–345. https://doi.org/10.1007/s00261-016-0950-1CrossRef

28.

Leng S, Takahashi N, Gomez Cardona D, Kitajima K, McCollough B, Li Z, Kawashima A, Leibovich BC, McCollough CH (2017) Subjective and objective heterogeneity scores for differentiating small renal masses using contrast-enhanced CT. Abdom Radiol N Y 42:1485–1492. https://doi.org/10.1007/s00261-016-1014-2CrossRef

29.

Lubner MG, Stabo N, Abel EJ, Del Rio AM, Pickhardt PJ (2016) CT Textural Analysis of Large Primary Renal Cell Carcinomas: Pretreatment Tumor Heterogeneity Correlates With Histologic Findings and Clinical Outcomes. AJR Am J Roentgenol 207:96–105. https://doi.org/10.2214/AJR.15.15451CrossRefPubMed

30.

Hodgdon T, McInnes MDF, Schieda N, Flood TA, Lamb L, Thornhill RE (2015) Can Quantitative CT Texture Analysis be Used to Differentiate Fat-poor Renal Angiomyolipoma from Renal Cell Carcinoma on Unenhanced CT Images? Radiology 276:787–796. https://doi.org/10.1148/radiol.2015142215CrossRefPubMed

31.

Scrima AT, Lubner MG, Abel EJ, Havighurst TC, Shapiro DD, Huang W, Pickhardt PJ (2018) Texture analysis of small renal cell carcinomas at MDCT for predicting relevant histologic and protein biomarkers. Abdom Radiol N Y. https://doi.org/10.1007/s00261-018-1649-2CrossRef

32.

Schieda N, Thornhill RE, Al-Subhi M, McInnes MDF, Shabana WM, van der Pol CB, Flood TA (2015) Diagnosis of Sarcomatoid Renal Cell Carcinoma With CT: Evaluation by Qualitative Imaging Features and Texture Analysis. AJR Am J Roentgenol 204:1013–1023. https://doi.org/10.2214/AJR.14.13279CrossRefPubMed

33.

Kim NY, Lubner MG, Nystrom JT, Swietlik JF, Abel EJ, Havighurst TC, Silverman SG, McGahan JP, Pickhardt PJ (2019) Utility of CT Texture Analysis in Differentiating Low-Attenuation Renal Cell Carcinoma From Cysts: A Bi-Institutional Retrospective Study. AJR Am J Roentgenol. https://doi.org/10.2214/AJR.19.21182CrossRefPubMedPubMedCentral

34.

Erickson BJ, Korfiatis P, Akkus Z, Kline TL (2017) Machine Learning for Medical Imaging. Radiogr Rev Publ Radiol Soc N Am Inc 37:505–515 . https://doi.org/10.1148/rg.2017160130

35.

He H, Ma Y (2013) Imbalanced learning: foundations, algorithms, and applications. John Wiley & Sons Inc, Hoboken, New JerseyCrossRef

36.

James G, Witten D, Hastie T, Tibshirani R (2013) An introduction to statistical learning: with applications in R. Springer, New YorkCrossRef

37.

DeLong ER, DeLong DM, Clarke-Pearson DL (1988) Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 44:837–845CrossRefPubMed

38.

Youden WJ (1950) Index for rating diagnostic tests. Cancer 3:32–35CrossRefPubMed

39.

Robin X, Turck N, Hainard A, Tiberti N, Lisacek F, Sanchez J-C, Müller M (2011) pROC: an open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinformatics 12:77. https://doi.org/10.1186/1471-2105-12-77CrossRefPubMedPubMedCentral

40.

Smith AD, Remer EM, Cox KL, Lieber ML, Allen BC, Shah SN, Herts BR (2012) Bosniak category IIF and III cystic renal lesions: outcomes and associations. Radiology 262:152–160. https://doi.org/10.1148/radiol.11110888CrossRefPubMed

41.

Smith AD, Allen BC, Sanyal R, Carson JD, Zhang H, Williams JH, Collins C, Griswold M, Zhang X (2015) Outcomes and complications related to the management of Bosniak cystic renal lesions. AJR Am J Roentgenol 204:W550–556. https://doi.org/10.2214/AJR.14.13149CrossRefPubMed

42.

Mousessian PN, Yamauchi FI, Mussi TC, Baroni RH (2017) Malignancy Rate, Histologic Grade, and Progression of Bosniak Category III and IV Complex Renal Cystic Lesions. AJR Am J Roentgenol 209:1285–1290. https://doi.org/10.2214/AJR.17.18142CrossRefPubMed

43.

Han W, Qin L, Bay C, Chen X, Yu K-H, Miskin N, Li A, Xu X, Young G (2020) Deep Transfer Learning and Radiomics Feature Prediction of Survival of Patients with High-Grade Gliomas. AJNR Am J Neuroradiol 41:40–48. https://doi.org/10.3174/ajnr.A6365CrossRefPubMedPubMedCentral

44.

Varghese BA, Cen SY, Hwang DH, Duddalwar VA (2019) Texture Analysis of Imaging: What Radiologists Need to Know. AJR Am J Roentgenol 212:520–528. https://doi.org/10.2214/AJR.18.20624CrossRefPubMed

45.

Foy JJ, Robinson KR, Li H, Giger ML, Al-Hallaq H, Armato SG (2018) Variation in algorithm implementation across radiomics software. J Med Imaging Bellingham Wash 5:044505. https://doi.org/10.1117/1.JMI.5.4.044505CrossRef

46.

Parekh V, Jacobs MA (2016) Radiomics: a new application from established techniques. Expert Rev Precis Med Drug Dev 1:207–226. https://doi.org/10.1080/23808993.2016.1164013CrossRefPubMedPubMedCentral

47.

Lee Y, Kim N, Cho K-S, Kang S-H, Kim DY, Jung YY, Kim JK (2009) Bayesian classifier for predicting malignant renal cysts on MDCT: early clinical experience. AJR Am J Roentgenol 193:W106–111. https://doi.org/10.2214/AJR.08.1858CrossRefPubMed

48.

Lubner MG, Smith AD, Sandrasegaran K, Sahani DV, Pickhardt PJ (2017) CT Texture Analysis: Definitions, Applications, Biologic Correlates, and Challenges. Radiogr Rev Publ Radiol Soc N Am Inc 37:1483–1503 . https://doi.org/10.1148/rg.2017170056

49.

Zhang G-M-Y, Shi B, Xue H-D, Ganeshan B, Sun H, Jin Z-Y (2019) Can quantitative CT texture analysis be used to differentiate subtypes of renal cell carcinoma? Clin Radiol 74:287–294. https://doi.org/10.1016/j.crad.2018.11.009CrossRefPubMed

50.

Bektas CT, Kocak B, Yardimci AH, Turkcanoglu MH, Yucetas U, Koca SB, Erdim C, Kilickesmez O (2019) Clear Cell Renal Cell Carcinoma: Machine Learning-Based Quantitative Computed Tomography Texture Analysis for Prediction of Fuhrman Nuclear Grade. Eur Radiol 29:1153–1163. https://doi.org/10.1007/s00330-018-5698-2CrossRefPubMed

51.

Haider MA, Vosough A, Khalvati F, Kiss A, Ganeshan B, Bjarnason GA (2017) CT texture analysis: a potential tool for prediction of survival in patients with metastatic clear cell carcinoma treated with sunitinib. Cancer Imaging Off Publ Int Cancer Imaging Soc 17:4. https://doi.org/10.1186/s40644-017-0106-8CrossRef

52.

Israel GM, Bosniak MA (2005) An update of the Bosniak renal cyst classification system. Urology 66:484–488. https://doi.org/10.1016/j.urology.2005.04.003CrossRefPubMed

53.

Israel GM, Bosniak MA (2005) How I do it: evaluating renal masses. Radiology 236:441–450. https://doi.org/10.1148/radiol.2362040218CrossRefPubMed

Title: Stratification of cystic renal masses into benign and potentially malignant: applying machine learning to the bosniak classification
Authors: Nityanand Miskin
Lei Qin
Shanna A. Matalon
Sree H. Tirumani
Francesco Alessandrino
Stuart G. Silverman
Atul B. Shinagare
Publication date: 01-01-2021
Publisher: Springer US
Keywords: Kidney Cancer
Kidney Cancer
Kidney Cancer
Published in: Abdominal Radiology / Issue 1/2021
Print ISSN: 2366-004X
Electronic ISSN: 2366-0058
DOI: https://doi.org/10.1007/s00261-020-02629-w

ACC 2024 Congress Coverage

Heart Failure Video

At a glance: The ONWARDS insulin icodec trials

Springer Medicine

Stratification of cystic renal masses into benign and potentially malignant: applying machine learning to the bosniak classification

Abstract

Purpose

Methods

Results

Conclusion

ACC 2024 Congress Coverage

Year in Review: Heart failure and cardiomyopathies

Semaglutide benefits people with type 2 diabetes and obesity-related heart failure

Incentivised to exercise

New intervention for STEMI-related cardiogenic shock

» View more highlights on the ACC 2024 congress hub page

At a glance: The ONWARDS insulin icodec trials

Springer Medicine

Abstract

Purpose

Methods

Results

Conclusion

Please log in to get access to this content

Other articles of this Issue 1/2021

Liver transplantation: current and future

Renal biopsies performed before versus during ablation of T1 renal tumors: implications for prevention of overtreatment and follow-up

CT and MR imaging evaluation of living liver donors

Sonography of liver transplantation

Identification and characterization of calyceal diverticula with MR urography (MRU) in children

Preoperative CT staging of borderline pancreatic cancer patients after neoadjuvant treatment: accuracy in the prediction of vascular invasion and resectability

ACC 2024 Congress Coverage

Year in Review: Heart failure and cardiomyopathies

Semaglutide benefits people with type 2 diabetes and obesity-related heart failure

Incentivised to exercise

New intervention for STEMI-related cardiogenic shock

» View more highlights on the ACC 2024 congress hub page