Abstract
Acute kidney injury (AKI), which refers to the abrupt decrease in renal function, is a common clinical condition in cardiovascular patients. The interaction between heart and kidney complicates the patient's condition, increasing the difficulty of treatment. Therefore, AKI is one of the focuses in the management of cardiovascular patients. Timely reversal of AKI is beneficial to the treatment of patients. In this study, we build a voting-based ensemble prediction model of acute kidney disease (AKD, referring to acute or subacute kidney damage for 7–90 days) risk for patients in coronary care units (CCU) based on clinical data within the first 6 h since CCU admission, in order to identify the patients whose AKI is hard to reverse. Then we interpret our prediction model using Shapley additive explanation values to help pinpoint important predictors and influence of clinical characteristics such as fluid status on AKD risk. The results show that our method has the potential to support clinical decision making.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Chawla, L.S., et al.: Acute kidney disease and renal recovery: consensus report of the acute disease quality initiative (ADQI) 16 Workgroup. Nat. Rev. Nephrol. 13(4), 241–257 (2017)
Uchino, S., Bellomo, R., Goldsmith, D., Bates, S., Ronco, C.: An assessment of the RIFLE criteria for acute renal failure in hospitalized patients. Crit. Care Med. 34(7), 1913–1917 (2006)
Susantitaphong, P., et al.: World incidence of AKI: a meta-analysis. Clin. J. Am. Soc. Nephrol. 8(9), 1482–1493 (2013)
Liano, F., Pascual, J.: Madrid Acute Renal Failure Study Group. Epidemiology of acute renal failure: a prospective, multicenter, community-based study. Kidney Int. 50(3), 811–818 (1996)
Vandenberghe, W., et al.: Acute kidney injury in cardiorenal syndrome type 1 patients: a systematic review and meta-analysis. Cardiorenal Med. 6(2), 116–128 (2015). https://doi.org/10.1159/000442300
Jentzer, J.C., et al.: Changes in comorbidities, diagnoses, therapies and outcomes in a contemporary cardiac intensive care unit population. Am. Heart J. 215, 12–19 (2019)
Deferrari, G., Cipriani, A., La Porta, E.: Renal dysfunction in cardiovascular diseases and its consequences. J. Nephrol. 34(1), 137–153 (2020)
Aronson, D., Burger, A.J.: The relationship between transient and persistent worsening renal function and mortality in patients with acute decompensated heart failure. J. Card. Fail. 16(7), 541–547 (2010)
Krishnamoorthy, A., et al.: Transient and persistent worsening renal function during hospitalization for acute heart failure. Am. Heart J. 168(6), 891–900 (2014)
Goldberg, A., Kogan, E., Hammerman, H., Markiewicz, W., Aronson, D.: The impact of transient and persistent acute kidney injury on long-term outcomes after acute myocardial infarction. Kidney Int. 76(8), 900–906 (2009)
Choi, J.S., et al.: Relation between transient or persistent acute kidney injury and long-term mortality in patients with myocardial infarction. Am. J. Cardiol. 112(1), 41–45 (2013)
Kellum, J.A., Sileanu, F.E., Bihorac, A., Hoste, E.A.J., Chawla, L.S.: Recovery after acute kidney injury. Am. J. Respir. Crit. Care Med. 195(6), 784–791 (2017)
Ronco, C., Ferrari, F., Ricci, Z.: Recovery after acute kidney injury: a new prognostic dimension of the syndrome. Am. J. Resp. Crit. Care Med. 195(6), 711–714 (2017)
Gameiro, J., Branco, T., Lopes, J.A.: Artificial intelligence in acute kidney injury risk prediction. J. Clin. Med. 9(3), 678 (2020)
Lin, K., Hu, Y., Kong, G.: Predicting in-hospital mortality of patients with acute kidney injury in the ICU using random forest model. Int. J. Med. Inform. 125, 55–61 (2019)
Chen, W., et al.: Causal risk factor discovery for severe acute kidney injury using electronic health records. BMC Med. Inform. Decis. Mak. 18(1), 13 (2018)
Ninet, S., Schnell, D., Dewitte, A., Zeni, F., Meziani, F., Darmon, M.: Doppler-based renal resistive index for prediction of renal dysfunction reversibility: a systematic review and meta-analysis. J. Crit. Care 30(3), 629–635 (2015)
Darmon, M., et al.: Diagnostic accuracy of Doppler renal resistive index for reversibility of acute kidney injury in critically ill patients. Intensive Care Med. 37(1), 68–76 (2011)
Brown, J.R., Kramer, R.S., MacKenzie, T.A., Coca, S.G., Sint, K., Parikh, C.R.: Determinants of acute kidney injury duration after cardiac surgery: an externally validated tool. Ann. Thorac. Surg. 93(2), 570–576 (2012)
Peerapornratana, S., Priyanka, P., Wang, S., et al.: Sepsis-associated acute kidney disease. Kidney Int. Rep. 5(6), 839–850 (2020)
Johnson, A.E.W., et al.: MIMIC-III, a freely accessible critical care database. Sci. data 3(1), 1–9 (2016)
Kellum, J.A., et al.: Kidney disease: improving global outcomes (KDIGO) acute kidney injury work group. KDIGO clinical practice guideline for acute kidney injury. Kidney Int. Suppl. 2(1), 1–138 (2012)
Breiman, L.: Random Forests. Mach. Learn. 2001 451, 45 (1), 5–32 (2001)
Chen, T., Guestrin, C.: XGBoost: A Scalable Tree Boosting System, In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (2016)
Ke, G., et al.: LightGBM: a highly efficient gradient boosting decision tree. In: Advances in Neural Information Processing Systems, vol. 30 (2017). https://github.com/Microsoft/LightGBM. Accessed 12 Jul 2021
He, H., Bai, Y., Garcia, E.A., S. Li.: ADASYN: Adaptive synthetic sampling approach for imbalanced learning. In: Proceedings of International Joint Conference on Neural Networks, pp. 1322–1328 (2008)
Lundberg, S., Lee, S.-I.: A Unified Approach to Interpreting Model Predictions. In: Advances in Neural Information Processing Systems, vol. 2017-Decem, pp. 4766–4775 (2017). http://arxiv.org/abs/1705.07874. Accessed 6 Sep 2020
Acknowledgements
The project is sponsored by Tsinghua-Toyota Joint Research Institute Inter-disciplinary Program.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Gong, K., Xie, X. (2021). An Interpretable Ensemble Model of Acute Kidney Disease Risk Prediction for Patients in Coronary Care Units. In: Qiu, R., Lyons, K., Chen, W. (eds) AI and Analytics for Smart Cities and Service Systems. ICSS 2021. Lecture Notes in Operations Research. Springer, Cham. https://doi.org/10.1007/978-3-030-90275-9_7
Download citation
DOI: https://doi.org/10.1007/978-3-030-90275-9_7
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-90274-2
Online ISBN: 978-3-030-90275-9
eBook Packages: Business and ManagementBusiness and Management (R0)