Prediction of Acute Burn-Induced Coagulopathy Risk with Machine Learning Models

Year 2023, Volume: 3 Issue: 2, 92 - 104, 01.10.2023

Murat Ali Çınar

Abstract

In burn patients, lipolysis, proteolysis, glycolysis and many severe hyperdynamic and hypermetabolic responses are seen with high fever. These hypermetabolic responses seen in patients lead to a decrease in lean muscle mass, delayed wound healing, weakening of the immune system, coagulopathy, and mortality. Particularly in the first 24 hours after major burns, fluid accumulation is observed in the interstitial space due to increased vascular permeability. Decreased intravascular volume affects tissue perfusion if not intervened. And the risk of coagulopathy increases due to acute burns. One of the most important causes of mortality in burn patients is seen as coagulopathy. Therefore, the application of machine learning-based decision support systems for rapid pre-diagnosis of coagulopathy may be important for clinicians and healthcare administrators.
In this study, machine learning models were investigated for the estimation of the risk of coagulopathy due to acute burns, using a data set of 1040 burn patients and 35 different biochemical parameters of these patients. The Subspace KNN model showed the highest prediction success compared to other machine learning methods with 100% accuracy.

Keywords

Burns, Coagulopathy, Machine Learning

References

• [1] Jeschke, M. G., van Baar, M. E., Choudhry, M. A., Chung, K. K., Gibran, N. S., Logsetty, S. (2020). Burn injury. Nature Reviews Disease Primers, 6(1), 1-25.
• [2] Greenhalgh, D. G. (2019). Management of burns. New England Journal of Medicine, 380(24), 2349- 2359.
• [3] Romero, S. A., Moralez, G., Jaffery, M. F., Huang, M., Cramer, M. N., Romain, N., Crandall, C. G. (2019). Progressive exercise training improves maximal aerobic capacity in individuals with well- healed burn injuries. American Journal of Physiology-Regulatory, Integrative and Comparative Physiology, 317(4), R563-R570.
• [4] Mandell, S. P., Gibran, N. S. (2014). Early enteral nutrition for burn injury. Advances in wound care, 3(1), 64-70.
• [5] Martini, W. Z., Holcomb, J. B., Yu, Y. M., Wolf, S. E., Cancio, L. C., Pusateri, A. E., Dubick, M. A. (2020). Hypercoagulation and Hypermetabolism of Fibrinogen in Severely Burned Adults. Journal of Burn Care & Research, 41(1), 23-29.
• [6] Williams, F. N., Jeschke, M. G., Chinkes, D. L., Suman, O. E., Branski, L. K., Herndon, D. N. (2009). Modulation of the hypermetabolic response to trauma: temperature, nutrition, and drugs. Journal of the American College of Surgeons, 208(4), 489-502.
• [7] Ball, R. L., Keyloun, J. W., Brummel-Ziedins, K., Orfeo, T., Palmieri, T. L., Johnson, L. S., Shupp, J. W. (2020). Burn-induced coagulopathies: a comprehensive review. Shock Augusta, Ga., 54(2), 154.
• [8] Glas, G. J., Levi, M., Schultz, M. J. (2016). Coagulopathy and its management in patients with severe burns. Journal of Thrombosis and Haemostasis, 14(5), 865-874.
• [9] Winter, W. E., Flax, S. D., Harris, N. S. (2017). Coagulation testing in the core laboratory. Laboratory Medicine, 48(4), 295-313.
• [10] Geng, K., Liu, Y., Yang, Y., Ding, X., Tian, X., Liu, H., Yan, H. (2020). Incidence and Prognostic Value of Acute Coagulopathy After Extensive Severe Burns. Journal of Burn Care & Research, 41(3), 544-549.
• [11] Lippi, G., Favaloro, E. J. (2008, October). Activated partial thromboplastin time: new tricks for an old dogma. In Seminars in thrombosis and hemostasis 34(07), (604-611).
• [12] Sherren, P. B., Hussey, J., Martin, R., Kundishora, T., Parker, M., & Emerson, B. (2013). Acute burn induced coagulopathy. Burns, 39(6), 1157-1161.
• [13] Mitra, B., Wasiak, J., Cameron, P. A., O’Reilly, G., Dobson, H., & Cleland, H. (2013). Early coagulopathy of major burns. Injury, 44(1), 40-43.
• [14] Zhu, F., Pan, Z., Tang, Y., Fu, P., Cheng, S., Hou, W., ... & Sun, Y. (2021). Machine learning models predict coagulopathy in spontaneous intracerebral hemorrhage patients in ER. CNS Neuroscience & Therapeutics, 27(1), 92-100.
• [15] Zhao, Q. Y., Liu, L. P., Luo, J. C., Luo, Y. W., Wang, H., Zhang, YJ, Luo, Z. (2021). A machine- learning approach for dynamic prediction of sepsis-induced coagulopathy in critically ill patients with sepsis. Frontiers in Medicine, 7, 637434.
• [16] Hasegawa, D., Yamakawa, K., Nishida, K., Okada, N., Murao, S., Nishida, O. (2020). Comparative analysis of three machine-learning techniques and conventional techniques for predicting sepsis-induced coagulopathy progression. Journal of clinical medicine, 9(7), 2113.
• [17] Li, K., Wu, H., Pan, F., Chen, L., Feng, C., Liu, Y., Li, T. (2020). A machine learning–based model to predict acute traumatic coagulopathy in trauma patients upon emergency hospitalization. Clinical and Applied Thrombosis/Hemostasis, 26, 1076029619897827.
• [18] Uddin, S., Khan, A., Hossain, M. E., Moni, M. A. (2019). Comparing different supervised machine learning algorithms for disease prediction. BMC medical informatics and decision making, 19(1), 1-16.
• [19] Choi, RY., Coyner, A. S., Kalpathy-Cramer, J., Chiang, M. F., Campbell, J. P. (2020). Introduction to machine learning, neural networks, and deep learning. Translational vision science & technology, 9(2), 14-14.
• [20] AKSOY, PK., Erdemir, F., KILINÇ, D., & Orhan, E. R. (2022). A Decision Support System on Artificial Intelligence Based Early Diagnosis of Sepsis. Artificial Intelligence Theory and Applications, 2(1), 14-26.
• [21] Safavian, S. R., and Landgrebe, D. (1991). A survey of decision tree classifier methodology. Institute of Electrical and Electronics Engineers Transactions on Systems, Man and Cybernetics, 21(3), 660-674.
• [22] Loh, W. Y. 82011). Classification and regression trees. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 1(1), 14-23.
• [23] Kavzoğlu, T. ve Çölkesen, İ. (20109. Karar ağaçları ile uydu görüntülerinin sınıflandırılması: Kocaeli örneği. Harita Teknolojileri Elektronik Dergisi, 2(1), 36-45.
• [24] Raj, R., Nehemiah, H. K., Elizabeth, D. S. and Kannan, A. (20189. A novel featuresignificance based k-nearest neighbour classification approach for computer aided diagnosis of lung disorders. Current Medical Imaging, 14(2), 289-300.
• [25] Fan, G. F., Guo, Y. H., Zheng, J. M. and Hong, W. C. (2019). Application of the weighted k-nearest neighbor algorithm for short-term load forecasting. Energies, 12(5), 916. 46
• [26] Jaber, M. M., Abd, S. K., Shakeel, P. M., Burhanuddin, M. A., Mohammed, M. A. And Yussof, S. (2020). A telemedicine tool framework for lung sounds classification using ensemble classifier algorithms. Measurement, 162, 107883.
• [27] Başer, F. ve Apaydın, A. (2015). Sınıflandırma amaçlı destek vektör makinelerinin lojistik regresyon ile karşılaştırılması. Anadolu University of Sciences & Technology-Theoretical Sciences, 3(2), 53-65.
• [28] Stehman, S. V. (1997). Selecting and interpreting measures of thematic classification accuracy. Remote sensing of Environment, 62(1), 77-89.
• [29] Çelik, Ö., Altunaydın, S. S. (2018). A Research on Machine Learning Methods and Its Applications. Online Learning, 1(3).