Akış madenciliği ve makine öğrenimi ile yoğun bakım hastalarında sepsis tahmini

Abstract

Çoklu organ yetmezliği olarak bilinen Sepsis hastalığı yoğun bakımlardaki tüm hastalar için, başka her ne hastalıklara sahip olurlarsa olsunlar, birinci mortalite sebebidir. Bu çalışmada yoğun bakım hastalarında sepsisi erken tahmin edebilen ve doktoru uyaran yoğun bakım ünitesi karar destek sistemi geliştirildi. COVID-19 virüsünün varyantı ve yoğun bakım hasta sayısı arttığından bu çalışma sepsis ile durumu kötüleştirmeye yönelik bir önlem olarak geliştirilmiştir. Hekimin hastanın sepsis durumunu daha iyi izlemesine yardımcı olmak için kullanıcı dostu bir arayüz ve sistem tasarlanmıştır. Türk hasta değerlerinin referans aralıklarına göre sepsis tahmini yapan bir karar destek sistemi ihtiyacını karşılamak amacıyla geliştirilmiştir. Sepsisi erken tahmin etmede daha iyi bir sonuç için, belirli bir süre içinde elde edilen ve kullanılan verilerin nasıl analiz edilmesi gerektiği ve daha yüksek performansı tahmin etmek için hangi yöntemlerin kullanılabileceği sonucuna varılmıştır. Çalışmada tahmin için makine öğrenmesi (sınıflandırma ve regresyon), derin öğrenme algoritmaları kullanılmış ve elde edilen sonuçlar karşılaştırılmıştır. Araştırmalar sonucunda, 300 yoğun bakım hastasına ait 122400 sa.’lik veriden oluşan ve Türk hastalarının referans aralıklarına göre yaklaşık %88 ile %94 arasında başarılı sonuçlar tahmin eden yoğun bakım sepsis karar destek sistemi geliştirilmiştir.

Keywords

• Derin öğrenme
• Karar destek sistemleri
• Makine öğrenmesi
• Tıbbi bilişim sistemleri
• Akış madenciliği

Prediction of sepsis for the intensive care unit patients with stream mining and machine learning

Abstract

Sepsis, which is known as multiple organ failure, is the primary cause of mortality for all patients in intensive care units, regardless of their other illnesses. An intensive care unit decision support system that can predict sepsis in intensive care patients early and warns the doctor has been developed. Since the COVID-19 virus, the variant and number of intensive care patients have increased, so this study has been developed as a precaution to worsen the situation with sepsis. A user-friendly interface and system have been designed to help the physician better monitor the patient's sepsis status. It has been developed in order to meet the need for a decision support system that makes sepsis estimation in accordance with the reference intervals of Turkish patients' values. For a better result of predicting sepsis early, it has been concluded how the data obtained and used in a certain period of time should be analyzed and what methods could be used to estimate higher performance. In the study, machine learning (classification and regression), deep learning algorithms have been used for estimation and the results obtained have been compared. As an impact of research, an intensive care sepsis decision support system, which consists of 122400 hourly data of 300 intensive care patients and estimates with approximately between 88% and 94% successful results in accordance with the reference intervals of Turkish patients, has been developed.

Keywords

• Deep learning
• Decision support systems
• Machine learning
• Medical information systems
• Stream mining

References

1. [1] Moschopoulos CD, Dimopoulou D, Dimopoulou A, Dimopoulou K, Protopapas K, Zavras N, Tsiodras S, Kotanidou A, Fragkou PC. “New Insights into the Fluid Management in Patients with Septic Shock”. Medicina, 59(6), 1-20, 2023.
2. [2] Rahmani K, Thapa R, Tsou P, Chetty SC, Barnes G, Lam C, Tso CF. “Assessing the effects of data drift on the performance of machine learning models used in clinical sepsis prediction”. International Journal of Medical Informatics, 173(1), 1-13, 2023.
3. [3] Bao C, Deng F, Zhao S. “Machine-learning models for prediction of sepsis patients mortality”. Medicina Intensiva (English Edition), 47(6), 315-325, 2023. World Health Organization. “Sepsis”. https://www.who.int/news-room/fact-sheets/detail/Sepsis (14.02.2023).
4. [4] Zhang TY, Zhong M, Cheng YZ, Zhang MW. “An interpretable machine learning model for real-time sepsis prediction based on basic physiological indicators”. European Review for Medical & Pharmacological Sciences, 27(10), 4348-4356, 2023.
5. [5] Dokuz Eylül Üniversitesi Araştırma Uygulama Hastanesi. “COVID-19 Pandemisi”. https://hastane.deu.edu.tr/ index.php/kurumsal.html?id=2698 (14.02.2023).
6. [6] Islam MM, Nasrin T, Walther BA, Wu CC, Yang HC, Li YC. “Prediction of sepsis patients using machine learning approach: a meta-analysis”. Computer Methods and Programs in Biomedicine, 170(1), 1-9, 2019.
7. [7] Kalantar B, Pradhan B, Naghibi SA, Motevalli A, Mansor S. “Assessment of the effects of training data selection on the landslide susceptibility mapping: a comparison between support vector machine, logistic regression and artificial neural networks. Geomatics”. Natural Hazards and Risk, 9(1), 49-69, 2018.
8. [8] van Wyk, F, Khojandi A, Kamaleswaran R. “Improving prediction performance using hierarchical analysis of real-time data: a sepsis case study”. IEEE journal of biomedical and health informatics, 23(3), 978-986, 2019.

9. [9] Tharwat A. “Classification assessment methods”. Applied Computing and Informatics, 17(1), 168-192, 2021.
10. [10] Saperstein Y, Ong SY, Al-Bermani T, Park J, Saperstein Y, Olayinka J, Jaiman A, Winer A, Salifu MO, McFarlane SI. “COVID-19 guidelines changing faster than the virus: implications of a clinical decision support app”. International Journal of Clinical Research & Trials, 5(2), 1-8, 2020.
11. [11] Wu G, Yang P, Xie Y, Woodruff HC, Rao X, Guiot J, Frix AN, Louis R, Moutschen M, Li J, Li J, Yan C, Du D, Zhao S, Ding Y, Liu B, Sun W, Albarello F, D’Abramo A, Schinià V, Nicastri E, Occhipinti M, Barisione G, Barisione E, Halilaj I, Lovinfosse P, Wang X, Wu J, Lambin P. “Development of a Clinical Decision Support System for Severity Risk Prediction and Triage of COVID-19 Patients at Hospital Admission: An International Multicenter Study”. European Respiratory Journal, 56(1), 1-11, 2020.
12. [12] Mandy J. “Arterial blood gas analysis. 1: Understanding ABG reports”. Nursing Times, 104(18), 28-29, 2008.
13. [13] Wernly B, Mamandipoor B, Baldia P, Jung C, Osmani V. “Machine learning predicts mortality in septic patients using only routinely available ABG variables: a multi-centre evaluation”. International Journal of Medical Informatics, 145(1), 1-9, 2020.
14. [14] Desautels T, Calvert J, Hoffman J, Jay M, Kerem Y, Shieh, L, Shimabukuro D, Chettipally U, Feldman MD, Barton C, Wales DJ, Das R. “Prediction of sepsis in the ICU with minimal electronic health record data: a machine learning approach”. JMIR medical informatics, 4(3), 1-15, 2016.
15. [15] Liu Z, Anahita K, Xueping L, Akram M, Robert LD, Rishikesan K. "A machine learning–enabled partially observable Markov decision process framework for early sepsis prediction." Informs Journal on Computing, 34(4), 2039-2057, 2002.
16. [16] Giannini HM, Ginestra JC, Chivers C, Draugelis M, Hanish A, Schweickert WD, Fuchs BD, Meadows L, Lynch M, Donnelly PJ, Pavan K, Fishman NO, Hanson W, Umscheid CA. “A machine learning algorithm to predict severe sepsis and septic shock: development, implementation, and impact on clinical practice”. Read Online: Critical Care Medicine| Society of Critical Care Medicine, 47(11), 1485-1492, 2019.
17. [17] Nemati S, Holder A, Razmi F, Stanley MD, Clifford GD, Buchman TG. “An interpretable machine learning model for accurate prediction of sepsis in the ICU”. Critical care medicine, 46(4), 547-553, 2018.
18. [18] McCoy A, Das R. “Reducing patient mortality, length of stay and readmissions through machine learning-based sepsis prediction in the emergency department, ICU and hospital floor units”. BMJ Open Quality, 6(2), 1-7, 2017.
19. [19] Shimabukuro DW, Barton CW, Feldman MD, Mataraso SJ, Das R. “Effect of a machine learning-based severe sepsis prediction algorithm on patient survival and hospital length of stay: a randomised clinical trial”. BMJ Open Respiratory Research. 4(1), 1-8, 2017.
20. [20] Barton C, Chettipally U, Zhou Y, Jiang Z, Lynn-Palevsky A, Le S, Calvert J, Das R. “Evaluation of a machine learning algorithm for up to 48-hour advance prediction of sepsis using six vital signs”. Computers in biology and medicine, 109(1), 79-84, 2019.
21. [21] Tran NK, Albahra S, Pham TN, Holmes JH, Greenhalgh D, Palmieri TL, Wajda J, Rashidi HH. “Novel application of an automated-machine learning development tool for predicting burn sepsis: proof of concept”. Scientific Reports, 10(1), 1-9, 2020.
22. [22] Ibrahim ZM, Wu H, Hamoud A, Stappen L, Dobson RJ, Agarossi A. “On classifying sepsis heterogeneity in the ICU: insight using machine learning”. Journal of the American Medical Informatics Association, 27(3), 437-443, 2020.
23. [23] Bayrak S, Doğan Y, Yılmaz R, Kut A. “ICU-Clinical decision support system”. Proc. 10th International Conferences Advances Semantic Process (SEMAPRO), Venice, Italy, 29 September-3 October 2016.
24. [24] Yoon J, Alaa A, Hu S, Schaar M. “ForecastICU: a prognostic decision support system for timely prediction of ICU admission”. International Conference on Machine Learning, New York City, NY, USA, 19-24 June 2016.
25. [25] Gupta A, Liu T, Shepherd S. “Clinical decision support system to assess the risk of sepsis using tree augmented Bayesian networks and electronic medical record data”. Health Informatics Journal, 26(2), 841-861, 2020.
26. [26] Amland RC, Haley JM, Lyons JJ. “A multidisciplinary sepsis program enabled by a two-stage clinical decision support system: factors that influence patient outcomes”. American Journal of Medical Quality, 31(6), 501-508, 2016.
27. [27] Fleuren LM, Klausch TL, Zwager CL, Schoonmade LJ, Guo T, Roggeveen LF, Swart EL, Girbes ARJ, Thoral P, Ercole A, Hoogendoorn M, Elbers PWG. “Machine learning for the prediction of sepsis: a systematic review and meta-analysis of diagnostic test accuracy”. Intensive Care Medicine, 46(1), 383-400, 2020.
28. [28] van Wyk F, Khojandi A, Kamaleswaran R. “Improving prediction performance using hierarchical analysis of real-time data: a sepsis case study”. IEEE Journal of Biomedical and Health Informatics, 23(3), 978-986, 2019.
29. [29] Meng C, Trinh L, Xu N, Enouen J, Liu Y. “Interpretability and fairness evaluation of deep learning models on MIMIC-IV dataset”. Scientific Reports, 12(1), 1-28, 2022.
30. [30] Hu W, Yang M, Chen H. “Database-based machine learning in sepsis deserves attention”. Intensive Care Medicine, 49(1), 262-263, 2023.
31. [31] Huyut MT, Üstündağ H. “Prediction of diagnosis and prognosis of COVID-19 disease by blood gas parameters using decision trees machine learning model: a retrospective observational study”. Medical Gas Research, 12(2), 60-68, 2022.
32. [32] Chen B, Maslove DM, Curran JD, Hamilton A, Laird PR, Mousavi P, Sibley S. “A deep learning model for the classification of atrial fibrillation in critically ill patients”. Intensive Care Medicine Experimental, 11(1), 1-10, 2023.
33. [33] Jentzer JC, Kashou AH, Murphree DH. “Clinical applications of artificial intelligence and machine learning in the modern cardiac intensive care unit”. Intelligence-Based Medicine, 7(1), 1-8, 2023.
34. [34] Dahmen J, Cook D. “SynSys: A synthetic data generation system for healthcare applications”. Sensors, 19(5), 1-11, 2019.
35. [35] Jiang X, Dai W, Cai Y. “Comparison of machine learning algorithms to SAPS II in predicting in-hospital mortality of fractures of the pelvis and acetabulum: analyzes based on The Medical Information Mart for Intensive Care (MIMIC)-III database”. All Life, 15(1), 1000-1012, 2022.
36. [36] Li XD, Li MM. “A novel nomogram to predict mortality in patients with stroke: a survival analysis based on the MIMIC-III clinical database”. BMC Medical Informatics and Decision Making, 22(1), 1-12, 2022.
37. [37] Khope SR, Elias S. “Strategies of predictive schemes and clinical diagnosis for prognosis using MIMIC-III: A systematic review”. Healthcare, 11(5), 1-24, 2023.