Öz
Giriş
İdrar yolu enfeksiyonları sık karşılaşılan bir tanı sorunudur. Altın standart olan idrar kültürü, hem zaman alıcıdır hem de çoğu zaman gereksiz yere istenir. Bu çalışmada, yalnızca yapılandırılmış laboratuvar verilerini kullanarak reflektif idrar kültürü istemine rehberlik edecek, yorumlanabilir bir makine öğrenimi (ML) tabanlı Laboratuvar Karar Destek Sistemi (LKDS) geliştirilmesi ve doğrulanması amaçlandı.
Gereç ve Yöntem
Retrospektif olarak 51.923 erişkin hastaya ait veriler incelendi. Yedi ML algoritması eğitildi; en yüksek doğruluk Rastgele Orman (Random Forest, RF) modelinde elde edildi. Model şeffaflığı için SHapley Additive exPlanations kullanıldı. En iyi 10 özellikten oluşan sadeleştirilmiş RF modeliyle üç farklı puanlama sistemi geliştirildi: Model doğruluğuna öncelik veren, tanısal dengeyi optimize eden ve hassasiyeti en üst düzeye çıkaran modeller.
Bulgular: RF modeli mükemmel performans gösterdi (harici testler – alıcı işletim karakteristiği eğrisi altında kalan alan [ROC-AUC]: 0,956). Basitleştirilmiş 10 değişkenli model yüksek doğruluğu korumuştur (ROC-AUC: 0,947). Temel öngörücüler arasında bakteri sayısı, lökositler, nitrit ve yaş yer almıştır. Skorlama sistemleri, farklı tanı hedeflerine göre uyarlanmış esnek seçenekler sunmuş ve SAFE-Skoru %95,3 hassasiyete ulaşmıştır.
Sonuç
Geliştirilen LKDS, gereksiz kültür sayısını azaltarak rasyonel antibiyotik kullanımını desteklemektedir. Açıklanabilir yapısı, laboratuvar profesyonelleriyle klinisyenler arasındaki iş birliğini kolaylaştırarak standartlaştırılmış reflektif test süreçlerine ve disiplinler arası karar vermeye katkı sağlar.
Anahtar Kelimeler:
İdrar yolu enfeksiyonları, makine öğrenimi, idrar kültürü
Kaynaklar
1
Schmiemann G, Kniehl E, Gebhardt K, Matejczyk MM, Hummers-Pradier E. The diagnosis of urinary tract infection: a systematic review. Dtsch Arztebl Int. 2010;107(21):361–7.
2
Flores E, Martínez-Racaj L, Blasco Á, Diaz E, Esteban P, López-Garrigós M, Salinas M. A step forward in the diagnosis of urinary tract infections: from machine learning to clinical practice. Comput Struct Biotechnol J. 2024;24:533–41.
3
O’Brien M, Marijam A, Mitrani-Gold FS, Terry L, Taylor-Stokes G, Joshi AV. Unmet needs in uncomplicated urinary tract infection in the United States and Germany: a physician survey. BMC Infect Dis. 2023;23:281.
4
Ozkan IA, Koklu M, Sert IU. Diagnosis of urinary tract infection based on artificial intelligence methods. Comput Methods Programs Biomed. 2018;166:51–9.
5
Choi MH, Kim D, Park Y, Jeong SH. Development and validation of artificial intelligence models to predict urinary tract infections and secondary bloodstream infections in adult patients. J Infect Public Health. 2024;17:10–7.
6
World Health Organization. Global Antimicrobial Resistance and Use Surveillance System (GLASS) report 2022. Geneva: World Health Organization; 2022.
7
Dedeene L, Van Elslande J, Dewitte J, Martens G, De Laere E, De Jaeger P, De Smet D. An artificial intelligence-driven support tool for prediction of urine culture test results. Clin Chim Acta. 2024;562:119854.
8
Devillé WL, Yzermans JC, van Duijn NP, Bezemer PD, van der Windt DA, Bouter LM. The urine dipstick test useful to rule out infections: a meta-analysis of the accuracy. BMC Urol. 2004;4:4.
9
Verboeket-van de Venne WPHG, Aakre KM, Watine J, Oosterhuis WP. Reflective testing: adding value to laboratory testing. Clin Chem Lab Med. 2012;50:1249–52.
10
Ramgopal S, Horvat CM, Yanamala N, Alpern ER. Machine learning to predict serious bacterial infections in young febrile infants. Pediatrics. 2020;146(3):e20194096.
11
Møller JK, Sørensen M, Hardahl C. Prediction of risk of acquiring urinary tract infection during hospital stay based on machine-learning: a retrospective cohort study. PLoS One. 2021;16:e0248636.
12
Herter WE, Khuc J, Cinà G, Knottnerus BJ, Numans ME, Wiewel MA, Bonten TN, de Bruin DP, van Esch T, Chavannes NH, Verheij RA. Impact of a machine learning–based decision support system for urinary tract infections: prospective observational study in 36 primary care practices. JMIR Med Inform. 2022;10:e27795.
13
Mancini A, Vito L, Marcelli E, Piangerelli M, De Leone R, Pucciarelli S, Merelli E. Machine learning models predicting multidrug resistant urinary tract infections using “DsaaS.” BMC Bioinformatics. 2020;21:347.
14
Shapiro Ben David S, Romano R, Rahamim-Cohen D, Azuri J, Greenfeld S, Gedassi B, Lerner U. AI driven decision support reduces antibiotic mismatches and inappropriate use in outpatient urinary tract infections. NPJ Digit Med. 2025;8:61.
15
Burton RJ, Albur M, Eberl M, Cuff SM. Using artificial intelligence to reduce diagnostic workload without compromising detection of urinary tract infections. BMC Med Inform Decis Mak. 2019;19:171.
16
Kranz J, Bartoletti R, Bruyère F, Cai T, Geerlings S, Köves B, Schubert S, Pilatz A, Veeratterapillay R, Wagenlehner FME, Bausch K, Devlies W, Horváth J, Leitner L, Mantica G, Mezei T, Smith EJ, Bonkat G. European Association of Urology Guidelines on urological infections: summary of the 2024 guidelines. Eur Urol. 2024;86(1):27–41.
17
Nelson Z, Aslan AT, Beahm NP, Blyth M, Cappiello M, Casaus D, Dominguez F, Egbert S, Hanretty A, Khadem T, Olney K, Abdul-Azim A, Aggrey G, Anderson DT, Barosa M, Bosco M, Chahine EB, Chowdhury S, Christensen A, de Lima Corvino D, Fitzpatrick M, Fleece M, Footer B, Fox E, Ghanem B, Hamilton F, Hayes J, Jegorovic B, Jent P, Jimenez-Juarez RN, Joseph A, Kang M, Kludjian G, Kurz S, Lee RA, Lee TC, Li T, Maraolo AE, Maximos M, McDonald EG, Mehta D, Moore WJ, Nguyen CT, Papan C, Ravindra A, Spellberg B, Taylor R, Thumann A, Tong SYC, Veve M, Wilson J, Yassin A, Zafonte V, Mena Lora AJ. Guidelines for the prevention, diagnosis, and management of urinary tract infections in pediatrics and adults: a WikiGuidelines Group consensus statement. JAMA Netw Open. 2024;7(11):e2444495. Erratum in: JAMA Netw Open. 2024;7(12):e2453497.
18
Werneburg GT, Lewis KC, Vasavada SP, Wood HM, Goldman HB, Shoskes DA, Li I, Rhoads DD. Urinalysis exhibits excellent predictive capacity for the absence of urinary tract infection. Urology. 2023;175:101–6.
19
Gharaghani M, Taghipour S, Halvaeezadeh M, Mahmoudabadi AZ. Candiduria: a review article with specific data from Iran. Turk J Urol. 2018;44(6):445–52.
20
Clinical and Laboratory Standards Institute (CLSI). Performance standards for antimicrobial susceptibility testing. 35th ed. CLSI document M100. Wayne (PA): CLSI; 2025.
21
Hosmer DW, Lemeshow S, Sturdivant RX. Applied logistic regression. 3rd ed. Hoboken (NJ): Wiley; 2013.
22
Yelin I, Snitser O, Novich G, Katz R, Tal O, Parizade M, Chodick G, Koren G, Shalev V, Kishony R. Personal clinical history predicts antibiotic resistance of urinary tract infections. Nat Med. 2019;25:1143–52.
23
Gadalla AAH, Friberg IM, Kift-Morgan A, Zhang J, Eberl M, Topley N, Weeks I, Cuff S, Wootton M, Gal M, Parekh G, Davis P, Gregory C, Hood K, Hughes K, Butler C, Francis NA. Identification of clinical and urine biomarkers for uncomplicated urinary tract infection using machine learning algorithms. Sci Rep. 2019;9:19694.
24
Werneburg GT, Rhoads DD, Milinovich A, McSweeney S, Knorr J, Mourany L, Zajichek A, Goldman HB, Haber GP, Vasavada SP. External validation of predictive models for antibiotic susceptibility of urine culture. BJU Int. 2025.
25
Hooton TM. Uncomplicated urinary tract infection. N Engl J Med. 2012;366:1028–37.
26
Foxman B. Epidemiology of urinary tract infections: incidence, morbidity, and economic costs. Am J Med. 2002;113:5–13.
27
de Vries S, ten Doesschate T, Totté JEE, Heutz JW, Loeffen YGT, Oosterheert JJ, Thierens D, Boel E. A semi-supervised decision support system to facilitate antibiotic stewardship for urinary tract infections. Comput Biol Med. 2022;146:105621.
28
Lin T-H, Chung H-Y, Jian M-J, Chang C-K, Lin H-H, Yu C-M, Perng C-L, Chang F-Y, Chen C-W, Chiu C-H, Shang H-S. Artificial intelligence-clinical decision support system for enhanced infectious disease management: accelerating ceftazidime-avibactam resistance detection in Klebsiella pneumoniae. J Infect Public Health. 2024;17:102541.
29
Dhanda G, Asham M, Shanks D, O’Malley N, Hake J, Satyan MT, Yedlinsky NT, Parente DJ. Adaptation and external validation of pathogenic urine culture prediction in primary care using machine learning. Ann Fam Med. 2023;21:11–8.
30
Taylor RA, Moore CL, Cheung K-H, Brandt C. Predicting urinary tract infections in the emergency department with machine learning. PLoS One. 2018;13:e0194085.
31
Seheult JN, Stram MN, Contis L, Pontzer RE, Hardy S, Wertz W, Baxter CM, Ondras M, Kip PL, Snyder GM, Pasculle AW. Development, evaluation, and multisite deployment of a machine learning decision tree algorithm to optimize urinalysis parameters for predicting urine culture positivity. J Clin Microbiol. 2023;61(6):e0029123.
32
Sergounioti A, Rigas D, Zoitopoulos V, Kalles D. From preliminary urinalysis to decision support: machine learning for UTI prediction in real-world laboratory data. J Pers Med. 2025;15(5):200.
33
Sheele JM, Campbell RL, Jones DD. Machine learning to predict bacteriuria in the emergency department. Sci Rep. 2025;15(1):31087.
34
Lachs MS, Nachamkin I, Edelstein PH, Goldman J, Feinstein AR, Schwartz JS. Spectrum bias in the evaluation of diagnostic tests: lessons from the rapid dipstick test for urinary tract infection. Ann Intern Med. 1992;117:135–40.
35
Zhao Q, Liu M, Gao K, Zhang B, Qi F, Xing T, Liu C, Gao J. Predicting 90-day risk of urinary tract infections following urostomy in bladder cancer patients using machine learning and explainability. Sci Rep. 2025;15:6807.
36
Wang H, Ding J, Wang S, Li L, Song J, Bai D. Enhancing predictive accuracy for urinary tract infections post-pediatric pyeloplasty with explainable AI: an ensemble TabNet approach. Sci Rep. 2025;15:2455.
37
Lee H-G, Seo Y, Kim JH, Han SB, Im JH, Jung CY, Durey A. Machine learning model for predicting ciprofloxacin resistance and presence of ESBL in patients with UTI in the ED. Sci Rep. 2023;13:3282.
38
Munigala S, Rojek R, Wood H, Yarbrough ML, Jackups RR, Burnham C-AD, Warren DK. Effect of changing urine testing orderables and clinician order sets on inpatient urine culture testing: analysis from a large academic medical center. Infect Control Hosp Epidemiol. 2019;40:281–86.
39
Fakih MG, Advani SD, Vaughn VM. Diagnosis of urinary tract infections: need for a reflective rather than reflexive approach. Infect Control Hosp Epidemiol. 2019;40:834–35.
40
Chambliss AB, Van TT. Revisiting approaches to and considerations for urinalysis and urine culture reflexive testing. Crit Rev Clin Lab Sci. 2022;59:112–24.
41
Chandrashekar G, Sahin F. A survey on feature selection methods. Comput Electr Eng. 2014;40:16-28.
42
Tonekaboni S, Joshi S, McCradden MD, Goldenberg A. What clinicians want: contextualizing explainable machine learning for clinical end use. Mach Learn Healthc. 2019:1–21.
43
Del Ben F, Da Col G, Cobârzan D, Turetta M, Rubin D, Buttazzi P, Antico A. A fully interpretable machine learning model for increasing the effectiveness of urine screening. Am J Clin Pathol. 2023;160:620–32.
44
General Directorate of Health Services (Türkiye Ministry of Health). Rational Laboratory Utilization Project (Akılcı Laboratuvar Kullanımı Projesi). Official Letter No. E.319. March 5, 2018. Ankara, Türkiye.
