Predicting of Bacteremia in Patients with Brucellosis Using Machine Learning Methods

Abstract

Purpose: The correct and early diagnosis of brucellosis is very crucial to decelerate its spread and providing fast treatment to patients. This study aims to develop a predictive model for diagnosing bacteremia in brucellosis patients based on some hematological and biochemical markers without the need for blood culture and bone marrow and to investigate the importance of these markers in predicting bacteremia. Materials/Methods: 162 patients with diagnosing brucellosis, 54.9% of whom are non-bacteremic, 45.1% bacteremia were retrospectively collected. The 20 demographic, hematological and biochemical laboratory parameters and 30 classifiers are used to predict bacteremia in brucellosis. Classifiers were developed by using Python programming language. Accuracy (ACC), Area under the receiver operating characteristic curve (AROC), and F measure were employed to find the best fit classification method. Feature importance method was used to determine most diagnostic markers to predict the bacteremia. Results: Extratree classifier with criterion “entropy” (ETC1) showed the best predictive performance with Acc values ranging between 0.5 and 1.00, F values between 0.53 and 1, and AROC values between 0.62 and 1. The neutrophil%, lymphocyte%, eosinophil%, alanine aminotransferase, and C-reactive protein were determined as the most distinguishing features with the scores 0.723, 1.000, 0.920, 0.869, and 0.769, respectively. Conclusions: This study showed that the ETC1 classifier may be helpful in determining bacteremia in brucellosis patients and that elevated lymphocytes, alanine aminotransferase, and C-reactive protein and low neutrophils and eosinophils may indicate bacteremic brucellosis.

Keywords

• Brucellosis
• Brucella
• Machine learning methods
• Classification
• Bacteremia

Brusellozlu Hastalarda Bakteriyeminin Makine Öğrenmesi Yöntemleri Kullanılarak Tahmin Edilmesi

Abstract

Amaç: Brusellozun doğru ve erken teşhisi, yayılımını yavaşlatmak ve hastalara hızlı tedavi sağlamak için çok önemlidir. Bu çalışmanın amacı, bruselloz hastalarında bakteriyemi tanısı için kan kültürü ve kemik iliğine kültürüne ihtiyaç duymadan bazı hematolojik ve biyokimyasal belirteçlere dayalı bir prediktif model geliştirmek ve bu belirteçlerin bakteriyemiyi öngörmedeki önemini araştırmaktır. Gereç/Yöntem: Bruselloz tanısı alan %54.9'u bakteriyemik olmayan, %45.1'i bakteriyemi olan 162 hasta retrospektif olarak toplandı. Brusellozda bakteriyemiyi öngörmek için 20 demografik, hematolojik ve biyokimyasal laboratuvar parametresi ve 30 sınıflandırıcı kullanılmıştır. Sınıflandırıcılar Python programlama dili kullanılarak geliştirilmiştir. En uygun sınıflandırma yöntemini bulmak için Doğruluk (ACC), Alıcı işletim karakteristik eğrisi altındaki alan (AROC) ve F ölçüsü kullanılmıştır. Bakteriyemiyi tahmin etmek için en tanısal belirteçleri belirlemek için özellik önemi yöntemi kullanılmıştır. Sonuçlar: "Entropi" ölçütlü ekstratree sınıflandırıcı (ETC1), 0,5 ile 1,00 arasında değişen Acc değerleri, 0,53 ile 1 arasında değişen F değerleri ve 0,62 ile 1 arasında değişen AROC değerleri ile en iyi tahmin performansını gösterdi. Nötrofil %, lenfosit %, eozinofil %, alanin aminotransferaz ve C-reaktif protein sırasıyla 0,723, 1,000, 0,920, 0,869 ve 0,769 skorlarıyla en ayırt edici özellikler olarak belirlenmiştir. Sonuçlar: Bu çalışma, ETC1 sınıflandırıcısının bruselloz hastalarında bakteriyemiyi belirlemede yardımcı olabileceğini, lenfosit, alanin aminotransferaz ve C-reaktif protein yüksekliğinin; nötrofil ve eozinofil düşüklüğünün bakteremik brusellozu gösterebileceğini göstermiştir.

Keywords

• Brucellosis
• Brucella
• Machine learning methods
• Classification
• Bacteremia

References

1. 1- Akhvlediani T, Bautista CT, Garuchava N, Sanodze L, Kokaia N, Malania L, et al. Epidemiological and clinical features of brucellosis in the country of Georgia. PLoS One 2017;12:e0170376. https://doi.org/10.1371/journal.pone.0170376
2. 2- Bahmani N, Bahmani A. A review of brucellosis in the Middle East and control of animal brucellosis in an Iranian experience. Reviews in Medical Microbiology 2022;33(1):e63-e69. doi: 10.1097/MRM.0000000000000266
3. 3- Yagupsky P, Morata P, Colmenero JD. Laboratory diagnosis of human brucellosis. Clinical Microbiology Reviews 2020;33(1):e00073-19. doi:10.1128/CMR.00073-19
4. 4- Buzgan T, Karahocagil MK, Irmak H, Baran AI, Karsen H, Evirgen O, et al. Clinical manifestations and complications in 1028 cases of brucellosis: a retrospective evaluation and review of the literature. International Journal of Infectious Diseases 2010;14(6):e469-478. https://doi.org/10.1016/j.ijid.2009.06.031
5. 5- Moosazadeh M, Nikaeen R, Abedi G, Kheradmand M, Safiri S. Epidemiological and clinical features of people with Malta fever in iran: a systematic review and meta-analysis. Osong Public Health and Research Perspectives 2016;7(3):157–167. https://doi.org/10.1016/j.phrp.2016.04.009
6. 6- Zheng R, Xie S, Lu X, Sun L, Zhou Y, Zhang Y, et al. A systematic review and meta-analysis of epidemiology and clinical manifestations of human brucellosis in China. BioMed research international 2018;2018:Article ID 5712920. https://doi.org/10.1155/2018/5712920
7. 7- Kadanali A, Ozden K, Altoparlak U, Erturk A, Parlak M. Bacteremic and nonbacteremic brucellosis: clinical and laboratory observations. Infection 2009;37(1):67-69. DOI 10.1007/s15010-008-7353-3
8. 8- Choudhury A, Kosorok MR. Missing data imputation for classication problems. arXiv:2002.10709 2020;1-27. https://doi.org/10.48550/arXiv.2002.10709

9. 9- Bailly A. Time Series Classification Algorithms with Applications in Remote Sensing. General Mathematics [math.GM]. Université Rennes 2, 2018. English.
10. 10- Shahub S., Upasham S., Ganguly A., Prasad S. Machine learning guided electrochemical sensors for passive sweat cortisol detection, Sensing and Bio-Sensing Research, 2022, 38,1-11, https://doi.org/10.1016/j.sbsr.2022.100527
11. 11- Breiman L. Bagging predictors. Machine learning 1996;24(2):123-140. https://doi.org/10.1007/BF00058655 12- Freund Y. Boosting a weak learning algorithm by majority. Information and computation 1995;121(2):256-285. https://doi.org/10.1006/inco.1995.1136
12. 13- Wolpert DH. Stacked generalization. Neural networks 1992;5(2):241-259. https://doi.org/10.1016/S0893-6080(05)80023-1
13. 14- Shahhosseini M, Hu G, Pham H. Optimizing ensemble weights and hyperparameters of machine learning models for regression problems. Machine Learning with Applications 2022;7:100251. https://doi.org/10.1016/j.mlwa.2022.100251
14. 15- Safdari R, Deghatipour A, Gholamzadeh M, Maghooli K. Applying data mining techniques to classify patients with suspected hepatitis C virus infection. Intelligent Medicine 2022;21:24. https://doi.org/10.1016/j.imed.2021.12.003
15. 16- Megahed A, Kandeel S, Alshaya DS, Attia KA, AlKahtani MD, Albohairy FM, et al. A comparison of logistic regression and classification tree to assess brucellosis associated risk factors in dairy cattle. Preventive Veterinary Medicine 2022;203:105664. https://doi.org/10.1016/j.prevetmed.2022.105664
16. 17- Al Dahouk S, Tomaso H, Nöckler K, Neubauer H, Frangoulidis D. Laboratory-based diagnosis of brucellosis – a review of the literature. Part I: techniques for direct detection and identification of Brucella spp. Clin Lab 2003;49(9–10):487–505.
17. 18- Al Dahouk S, Nöckler K. Implications of laboratory diagnosis on brucellosis therapy. Expert Review of Anti-infective Therapy 2011;9(7):833-845, https://doi.org/10.1586/eri.11.55
18. 19- Pappas G, Papadimitriou P. Challenges in Brucella bacteraemia. International Journal of Antimicrobial Agents 2007;30:29-31. doi:10.1016/j.ijantimicag.2007.06.011
19. 20- Qie C, Cui J, Liu Y, Li Y, Wu H, Mi Y. Epidemiological and clinical characteristics of bacteremic brucellosis. Journal of International Medical Research 2020;48(7):1-7. doi:10.1177/0300060520936829
20. 21- Özdem S, Tanır G, Öz FN, Yalçınkaya R, Cinni RG, Şen ZS, et al. Bacteremic and Nonbacteremic Brucellosis in Children in Turkey. Journal of Tropical Pediatrics 2022;68(1): 114. https://doi.org/10.1093/tropej/fmab114
21. 22- Kara SS, Cayir Y. Predictors of blood culture positivity in pediatric brucellosis. J Coll Physicians Surg Pak 2019;29(07):665-670.
22. 23- Chicco D, Jurman G. An ensemble learning approach for enhanced classification of patients with hepatitis and cirrhosis. IEEE Access, 2021; 9:24485-98. doi: 10.1109/ACCESS.2021.3057196.
23. 24- Chicco D, Oneto L. Data analytics and clinical feature ranking of medical records of patients with sepsis. BioData Mining 2021;14(12):1-22.
24. 25- Xiong Y, Ma Y, Ruan L, Li D, Lu C, Huang L. Comparing different machine learning techniques for predicting COVID-19 severetity. Infectious Diseases of Poverty 2022;11(1):1-9. https://doi.org/10.1186/s40249-022-00946-4 26- Kou Z, Fan X, Li J, Shao Z, Qiang X. Using amino acid features to identify the pathogenicity of influenza B virüs. Infectious Diseases of Poverty 2022;11(1):1-13 https://doi.org/10.1186/s40249-022-00974-0.