Predictive Value of Machine Learning Model Based on CT Values for Urinary Tract Infection Stones

Overview

Journal iScience

Publisher Cell Press

Date 2024 Dec 5

PMID 39634558

Authors

Jiaxin Li

Yao Du

Gaoming Huang

Chiyu Zhang

Zhenfeng Ye

Jinghui Zhong

Xiaoqing Xi

Yawei Huang

Affiliations

Soon will be listed here.

Abstract

Preoperative diagnosis of infection stones presents a significant clinical challenge. We developed a machine learning model to predict urinary infection stones using computed tomography (CT) values, enabling preoperative identification. In this study, we included 1209 patients who underwent urinary lithotripsy at our hospital. Seven machine learning algorithms along with eleven preoperative variables were used to construct the prediction model. Subsequently, model performance was evaluated by calculating AUC and AUPR for subjects in the validation set. On the validation set, all seven machine learning models demonstrated strong discrimination (AUC: 0.687-0.947). Additionally, the XGBoost model was identified as the optimal model significantly outperforming the traditional LR model. Taken together, the XGBoost model is the first machine learning model for preoperative prediction of infection stones based on CT values. It can rapidly and accurately identify infection stones , providing valuable guidance for urologists in managing these stones.

References

Marchini G, Gebreselassie S, Liu X, Pynadath C, Snyder G, Monga M . Absolute Hounsfield unit measurement on noncontrast computed tomography cannot accurately predict struvite stone composition. J Endourol. 2012; 27(2):162-7. DOI: 10.1089/end.2012.0470. View

Bichler K, Eipper E, Naber K, Braun V, Zimmermann R, Lahme S . Urinary infection stones. Int J Antimicrob Agents. 2002; 19(6):488-98. DOI: 10.1016/s0924-8579(02)00088-2. View

Harmon W, Sershon P, Blute M, Patterson D, Segura J . Ureteroscopy: current practice and long-term complications. J Urol. 1997; 157(1):28-32. DOI: 10.1016/s0022-5347(01)65272-8. View

Talati V, Soares R, Khambati A, Nadler R, Perry Jr K . Trends in urinary calculi composition from 2005 to 2015: a single tertiary center study. Urolithiasis. 2019; 48(4):305-311. DOI: 10.1007/s00240-019-01151-z. View

Gonzalez R, Whiting B, Canales B . The history of kidney stone dissolution therapy: 50 years of optimism and frustration with renacidin. J Endourol. 2011; 26(2):110-8. PMC: 3311908. DOI: 10.1089/end.2011.0380. View

Shen J, Xiao Z, Wang X, Zhao Y . A nomogram clinical prediction model for predicting urinary infection stones: development and validation in a retrospective study. World J Urol. 2024; 42(1):211. DOI: 10.1007/s00345-024-04904-7. View

Gadalla A, Friberg I, Kift-Morgan A, Zhang J, Eberl M, Topley N . Identification of clinical and urine biomarkers for uncomplicated urinary tract infection using machine learning algorithms. Sci Rep. 2019; 9(1):19694. PMC: 6928162. DOI: 10.1038/s41598-019-55523-x. View

Alkadri S, Ledwos N, Mirchi N, Reich A, Yilmaz R, Driscoll M . Utilizing a multilayer perceptron artificial neural network to assess a virtual reality surgical procedure. Comput Biol Med. 2021; 136:104770. DOI: 10.1016/j.compbiomed.2021.104770. View

Koras O, Bozkurt I, Yonguc T, Degirmenci T, Arslan B, Gunlusoy B . Risk factors for postoperative infectious complications following percutaneous nephrolithotomy: a prospective clinical study. Urolithiasis. 2014; 43(1):55-60. DOI: 10.1007/s00240-014-0730-8. View

10.

Knoll T, Schubert A, Fahlenkamp D, Leusmann D, Wendt-Nordahl G, Schubert G . Urolithiasis through the ages: data on more than 200,000 urinary stone analyses. J Urol. 2011; 185(4):1304-11. DOI: 10.1016/j.juro.2010.11.073. View

11.

Kulkarni N, Eisner B, Pinho D, Joshi M, Kambadakone A, Sahani D . Determination of renal stone composition in phantom and patients using single-source dual-energy computed tomography. J Comput Assist Tomogr. 2013; 37(1):37-45. DOI: 10.1097/RCT.0b013e3182720f66. View

12.

Wu Y, Mo Q, Xie Y, Zhang J, Jiang S, Guan J . A retrospective study using machine learning to develop predictive model to identify urinary infection stones in vivo. Urolithiasis. 2023; 51(1):84. PMC: 10232574. DOI: 10.1007/s00240-023-01457-z. View

13.

Martinez-Pineiro J, de Iriarte E, Armero A . The problem of recurrences and infection after surgical removal of staghorn calculi. Eur Urol. 1982; 8(2):94-101. DOI: 10.1159/000473488. View

14.

Barah M, Mehrotra S . Predicting Kidney Discard Using Machine Learning. Transplantation. 2021; 105(9):2054-2071. PMC: 8263801. DOI: 10.1097/TP.0000000000003620. View

15.

Kalafi E, Nor N, Taib N, Ganggayah M, Town C, Dhillon S . Machine Learning and Deep Learning Approaches in Breast Cancer Survival Prediction Using Clinical Data. Folia Biol (Praha). 2020; 65(5-6):212-220. DOI: 10.14712/fb2019065050212. View

16.

Beck E, Riehle Jr R . The fate of residual fragments after extracorporeal shock wave lithotripsy monotherapy of infection stones. J Urol. 1991; 145(1):6-9; discussion 9-10. DOI: 10.1016/s0022-5347(17)38230-7. View

17.

Singh M, Chapman R, TRESIDDER G, Blandy J . The fate of the unoperated staghorn calculus. Br J Urol. 1973; 45(6):581-5. DOI: 10.1111/j.1464-410x.1973.tb12226.x. View

18.

Tiselius H, Andersson A . Stone burden in an average Swedish population of stone formers requiring active stone removal: how can the stone size be estimated in the clinical routine?. Eur Urol. 2003; 43(3):275-81. DOI: 10.1016/s0302-2838(03)00006-x. View

19.

Nevo A, Golomb D, Lifshitz D, Yahav D . Predicting the risk of sepsis and causative organisms following urinary stones removal using urinary versus stone and stent cultures. Eur J Clin Microbiol Infect Dis. 2019; 38(7):1313-1318. DOI: 10.1007/s10096-019-03555-6. View

20.

Lieske J, de la Vega L, Gettman M, Slezak J, Bergstralh E, Melton 3rd L . Diabetes mellitus and the risk of urinary tract stones: a population-based case-control study. Am J Kidney Dis. 2006; 48(6):897-904. DOI: 10.1053/j.ajkd.2006.09.002. View