Interpretation of the prognosis of early results of nephron-sparing surgery with consideration of surgical learning curve using clinical decision support systems

Cover Page

Cite item

Full Text

Open Access Open Access
Restricted Access Access granted
Restricted Access Subscription or Fee Access

Abstract

Aim. To assess the possibility of interpreting machine learning models to predict the early results of laparoscopic nephron-sparing surgery (NSS) in kidney tumors with consideration of surgical learning curve.

Materials and methods. The results of 320 consecutive laparoscopic NSS in patients with localized kidney tumors, performed by 4 surgeons, were analyzed. The construction of a machine learning model taking into account surgical learning curve was carried out based on the extreme gradient boosting (eXtreme Gradient Boosting). To identify significant factors and interpret the prognostic ability of the model, the SHapley Additive exPlanations method was used with a calculation of the Shapley value. Three groups of factors were chosen as an array of input data. The first group included demographic and clinical characteristics of patients, such as age, gender, Charlson comorbidity index, body mass index, preoperative glomerular filtration rate (GFR). In the second group, there were morphometric indicators of the kidney tumor, including RENAL. Nephrometry Score, PADUA (Preoperative Aspects and Dimensions Used for an Anatomical), C-index (Centrality index score), absolute tumor volume, localization of the tumor in relation to the kidney surface. In addition, factors associated with surgical learning curve, such as case number and perioperative results last 10 procedures, were analyzed. The target variables were duration of the procedure, warm ischemia time, and postoperative GFR after 24 hours.

Results. The SHAP method allows a visual interpretation of a machine learning algorithm based on the extreme gradient boosting for individual prediction of early perioperative outcomes of laparoscopic NSS in patients with renal tumors. For the calculated new features “complexity”, “slope angle” and others using the SHAP method, the high significance in building predictive models for target variables was confirmed, and an interpretation of the influence of specific features on the target variable in the constructed machine learning models was also given.

Conclusion. The SHAP method showed good practical results that coincide with the observations of specialists. The use of such solutions will allow in the future to introduce machine learning models to form clinical decision support systems.

Full Text

Restricted Access

About the authors

E. S. Syrota

Institute of Urology and Reproductive Health, FGAOU VO I.M. Sechenov First Moscow State Medical University; FGBUN Center for Information Technologies in Design, Russian Academy of Sciences

Author for correspondence.
Email: sirota_e_s@staff.sechenov.ru

Ph.D. in technical sciences, associate professor, Head of the Laboratory 

Russian Federation, Moscow; Odintsovo

I. A. Kuznetsov

FGBUN Center for Information Technologies in Design, Russian Academy of Sciences

Email: info@ditc.ras.ru

Ph.D. in technical sciences, associate professor, Head of the Laboratory 

Russian Federation, Odintsovo

P. V. Glybochko

Institute of Urology and Reproductive Health, FGAOU VO I.M. Sechenov First Moscow State Medical University

Email: glybochko_p_v@staff.sechenov.ru

academician of RSA, professor, Ph.D., MD, rector

Russian Federation, Moscow

D. V. Butnaru

Institute of Urology and Reproductive Health, FGAOU VO I.M. Sechenov First Moscow State Medical University

Email: butnaru_dv@mail.ru

M.D., urologist, associate professor, Chief Physician of the University Clinical Hospital No. 1 

Russian Federation, Moscow

Yu. G. Alyaev

Institute of Urology and Reproductive Health, FGAOU VO I.M. Sechenov First Moscow State Medical University

Email: ugalyaev@mail.ru

corresponding member of RAS, Ph.D., MD, professor 

Russian Federation, Moscow

D. N. Fiev

Institute of Urology and Reproductive Health, FGAOU VO I.M. Sechenov First Moscow State Medical University

Email: fiev_d_n@staff.sechenov.ru

Ph.D., MD, urologist, chief researcher 

Russian Federation, Moscow

A. V. Proskura

Institute of Urology and Reproductive Health, FGAOU VO I.M. Sechenov First Moscow State Medical University

Email: proskura_a_v_1@staff.sechenov.ru

Ph.D., urologist, oncologist, assistant 

Russian Federation, Moscow

A. R. Adzhiev

Institute of Urology and Reproductive Health, FGAOU VO I.M. Sechenov First Moscow State Medical University

Email: adzhiev-1998@bk.ru

2-year resident

Russian Federation, Moscow

A. A. Zholdubaev

Institute of Urology and Reproductive Health, FGAOU VO I.M. Sechenov First Moscow State Medical University

Email: gradmonaco@yandex.ru

urologist, Ph.D. student 

Russian Federation, Moscow

References

  1. Kaprin A.D., Starinsky V.V., Shakhzadova A.O. The state of oncological care for the population of Russia in 2022. M.: P.A. Herzen Moscow State Medical Research Institute − branch of the Federal State Budgetary Institution «NMIC of Radiology» of the Ministry of Health of Russia, 2022. fig. 239 p. Russian (Каприн А.Д., Старинский В.В., Шахзадова А.О. Состояние онкологической помощи населению России в 2022 году. М.: МНИОИ им. П.А. Герцена − филиал ФГБУ «НМИЦ радиологии» Минздрава России, 2022. илл. 239 с.).
  2. Bukavina L., Bensalah K., Bray F., Carlo M., Challacombe B., Karam J.A., Kassouf W., Mitchell T., Montironi R., O’Brien T, Panebianco V., Scelo G., Shuch B., van Poppel H., Blosser C.D., Psutka S.P. Epidemiology of Renal Cell Carcinoma: 2022 Update. Eur Urol. 2022; 82(5):529–542. doi: 10.1016/j.eururo.2022.08.019.
  3. Vasudev N.S., Wilson M., Stewart G.D., et al Challenges of early renal cancer detection: symptom patterns and incidental diagnosis rate in a multicentre prospective UK cohort of patients presenting with suspected renal cancer. BMJ Open. 2020;10:e035938. doi: 10.1136/bmjopen-2019-035938.
  4. Ljungberg B., Albiges L., Abu-Ghanem Y., Bedke J., Capitanio U., Dabestani S., Fernández-Pello S., Giles R.H., Hofmann F., Hora M., Klatte T., Kuusk T., Lam T.B., Marconi L., Powles T., Tahbaz R., Volpe A., Bex A. European Association of Urology Guidelines on Renal Cell Carcinoma: The 2022 Update. Eur Urol. 2022;82(4):399–410. doi: 10.1016/j.eururo.2022.03.006.
  5. Campbell R.A., Scovell J., Rathi N., Aram P., Yasuda Y., Krishnamurthi V., Eltemamy M., Goldfarb D., Wee A., Kaouk J., Weight C., Haber G.P., Campbell S.C. Partial Versus Radical Nephrectomy: Complexity of Decision-Making and Utility of AUA Guidelines. Clin Genitourin Cancer. 2022;20(6):501–509. doi: 10.1016/j.clgc.2022.06.003.
  6. Clinical recommendations – Cancer of the renal parenchyma. 2021–2022–2023 (01/20/2023) – Approved by the Ministry of Health of the Russian Federation http://disuria.ru/load/zakonodatelstvo/klinicheskie_rekomendacii_protokoly_lechenija/54. Russian (Клинические рекомендации – Рак паренхимы почки. 2021–2022–2023 (20.01.2023) – Утверждены Минздравом РФ http://disuria.ru/load/zakonodatelstvo/klinicheskie_rekomendacii_protokoly_lechenija/54)
  7. Castelvecchi D. Can we open the black box of AI? Nature News. 2016;538(7623):20.
  8. Cacciamani G.E., Gill T., Medina L., Ashrafi A., Winter M., Sotelo R., Artibani W., Gill I.S. Impact of Host Factors on Robotic Partial Nephrectomy Outcomes: Comprehensive Systematic Review and Meta-Analysis. J Urol. 2018;200(4):716–730. doi: 10.1016/j.juro.2018.04.079.
  9. Shapley L.S. et al. A value for n-person games. 1953.
  10. Alyaev Yu.G., Sirota E.S., Bezrukov E.A., Sukhanov R.B. Computer-assisted laparoscopic operations in the surgical treatment of kidney cancer. Urology. 2018;3:30–38. Russian (Аляев Ю.Г., Сирота Е.С., Безруков Е.А., Суханов Р.Б. Компьютер-ассистированные лапароскопические операции при хирургическом лечении рака почки. Урология. 2018;3:30–38).
  11. Gridin V.N., Kuznetsov I.A., Gazov A.I., Sirota E.S. Prediction of perioperative parameters of laparoscopic organ-preserving interventions on the kidney taking into account the «learning curve» of the surgeon. Biomedical radio electronics. 2021;24(2):13–20. Day: https://doi.org/10.18127/j15604136-202102-02. Russian (Гридин В.Н., Кузнецов И.А., Газов А.И., Сирота Е.С. Прогнозирование периоперационных параметров лапароскопических органосохранных вмешательств на почке с учетом «кривой обучения» хирурга. Биомедицинская радиоэлектроника. 2021;24(2):13–20. Doi: https://doi.org/10.18127/j15604136-202102-02).
  12. Fero K., Hamilton Z.A., Bindayi A., Murphy J.D., Derweesh I.H. Utilization and quality outcomes of cT1a, cT1b and cT2a partial nephrectomy: analysis of the national cancer database. BJU Int. 2018;121(4):565–574. doi: 10.1111/bju.14055.
  13. MacLennan S. et al. Systematic review of oncological outcomes following surgical management of localised renal cancer. Eur. Urol. 2012;61(5):972–993. doi: 10.1016/j.eururo.2012.02.039.
  14. Eden C.G., Arora A., Hutton A. Cancer control, continence, and potency after laparoscopic radical prostatectomy beyond the learning and discovery curves. J Endourol. 2011;25(5):815–819. doi: 10.1089/end.2010.0451.
  15. Alimi Q., Peyronnet B., Sebe P., Cote J.F., Kammerer-Jacquet S.F., Khene Z.E., Pradere B., Mathieu R., Verhoest G., Guillonneau B., Bensalah K. Comparison of Short-Term Functional, Oncological, and Perioperative Outcomes Between Laparoscopic and Robotic Partial Nephrectomy Beyond the Learning Curve. J Laparoendosc Adv Surg Tech A. 2018;28(9):1047–1052. doi: 10.1089/lap.2017.0724.
  16. Abboudi H., Khan M.S., Guru K.A., Froghi S., de Win G., Van Poppel H., Dasgupta P., Ahmed K. Learning curves for urological procedures: a systematic review. BJU Int. 2014;114(4):617–629. doi: 10.1111/bju.12315.
  17. Sirota E.S., Rapoport L.M., Gridin V.N., Tsarichenko D.G., Kuznetsov I.A., Sirota A.E, Alyaev Y.G. Analysis of the learning curve in laparoscopic partial nephrectomy in patients with localized renal parenchymal lesions depending on the nephrometric score. Urologiia. 2020;(6):11–18. Russian. doi: 10.18565/urology.2020.6.11-18.
  18. Zhang Z., Beck M.W., Winkler D.A., Huang B., Sibanda W., Goyal H.; written on behalf of AME Big-Data Clinical Trial Collaborative Group. Opening the black box of neural networks: methods for interpreting neural network models in clinical applications. Ann Transl Med. 2018;6(11):216. doi: 10.21037/atm.2018.05.32.
  19. Buzink S.N., Botden S.M., Heemskerk J., Goossens R.H., de Ridder H., Jakimowicz J.J. Camera navigation and tissue manipulation; are these laparoscopic skills related? Surg Endosc. 2009;23(4):750–757. doi: 10.1007/s00464-008-0057-z.
  20. Lee S.M., Robertson I., Stonier T., Simson N., Amer T., Aboumarzouk O.M. Contemporary outcomes and prediction of adherent perinephric fat at partial nephrectomy: a systematic review. Scand J Urol. 2017;51(6):429–434. doi: 10.1080/21681805.2017.1357656.
  21. Thompson R.H., Lane B.R., Lohse C.M., Leibovich B.C., Fergany A., Frank I., Gill I.S,. Blute M.L., Campbell S.C. Every minute counts when the renal hilum is clamped during partial nephrectomy. Eur Urol. 2010;58(3):340–345. doi: 10.1016/j.eururo.2010.05.047.
  22. Simone G., Gill I.S., Mottrie A., Kutikov A., Patard J.-J., Alcaraz A., et al. Indications, techniques, outcomes, and limitations for minimally ischemic and off clamp partial nephrectomy: a systematic review of the literature. Eur Urol. 2015;68(4):632–640. doi: 10.1016/j.eururo.2015.04.020.
  23. Bianchi L., Schiavina R., Bortolani B., Cercenelli L., Gaudiano C., Mottaran A., Droghetti M., Chessa F., Boschi S., Molinaroli E., Balestrazzi E., Costa F., Rustici A., Carpani G., Piazza P., Cappelli A., Bertaccini A., Golfieri R., Marcelli E., Brunocilla E. Novel Volumetric and Morphological Parameters Derived from Three-dimensional Virtual Modeling to Improve Comprehension of Tumor’s Anatomy in Patients with Renal Cancer. Eur Urol Focus. 2022;8(5):1300–1308. doi: 10.1016/j.euf.2021.08.002.
  24. Hu C., Sun J., Zhang Z., Zhang H., Zhou Q., Xu J., Ling Z., Ouyang J. Parallel comparison of R.E.N.A.L., PADUA, and C-index scoring systems in predicting outcomes after partial nephrectomy: A systematic review and meta-analysis. Cancer Med. 2021;10(15):5062–5077. doi: 10.1002/cam4.4047.
  25. Veccia A., Antonelli A., Uzzo R.G., Novara G., Kutikov A., Ficarra V., Simeone C., Mirone V., Hampton L.J., Derweesh I., Porpiglia F., Autorino R. Predictive Value of Nephrometry Scores in Nephron-sparing Surgery: A Systematic Review and Meta-analysis. Eur Urol Focus. 2020;6(3):490–504. doi: 10.1016/j.euf.2019.11.004.
  26. Mir M.C., Ercole C., Takagi T., Zhang Z., Velet L., Remer E.M., Demirjian S., Campbell S.C. Decline in renal function after partial nephrectomy: etiology and prevention. J Urol. 2015;193(6):1889–1898. doi: 10.1016/j.juro.2015.01.093.
  27. Bravi C.A., Vertosick E., Benfante N., Tin A., Sjoberg D,. Hakimi A.A, Touijer K., Montorsi F., Eastham J., Russo P., Vickers A. Impact of Acute Kidney Injury and Its Duration on Long-term Renal Function After Partial Nephrectomy. Eur Urol. 2019;76(3):398–403. doi: 10.1016/j.eururo.2019.04.040.
  28. Campbell S.C. A nonischemic approach to partial nephrectomy is optimal. No. J Urol. 2012;187(2):388–90.
  29. Volpe A., Blute M.L., Ficarra V., Gill I.S., Kutikov A., Porpiglia F., Rogers C., Touijer K.A., Van Poppel H., Thompson R.H. Renal Ischemia and Function After Partial Nephrectomy: A Collaborative Review of the Literature. Eur Urol. 2015;68(1):61–74. doi: 10.1016/j.eururo.2015.01.025.

Supplementary files

Supplementary Files
Action
1. JATS XML
2. Fig. 1. Shepley's value for parameters by duration of surgical intervention

Download (489KB)
3. Fig. 2. Shepley's value for parameters on the time of thermal ischemia

Download (433KB)
4. Fig. 3. Shepley's value for parameters on glomerular filtration rate after 24 hours

Download (262KB)

Copyright (c) 2024 Bionika Media

This website uses cookies

You consent to our cookies if you continue to use our website.

About Cookies