Machine Learning-Assisted Ensemble Analysis for the Prediction of Acute Pancreatitis with Acute Kidney Injury

Abstract

Purpose Acute kidney injury (AKI) is a frequent complication of severe acute pancreatitis (AP) and carries a very poor prognosis. The present study aimed to construct a model capable of accurately identifying those patients at high risk of harboring occult acute kidney injury (AKI) characteristics. Patients and Methods We retrospectively recruited a total of 424 consecutive patients at the Gezhouba central hospital of Sinopharm and Xianning central hospital between January 1, 2016, and October 30, 2021. ML-assisted models were developed from candidate clinical features using two-step estimation methods. The receiver operating characteristic curve (ROC), decision curve analysis (DCA), and clinical impact curve (CIC) were performed to evaluate the robustness and clinical practicability of each model. Results Finally, a total of 30 candidate variables were included, and the AKI prediction model was established by an ML-based algorithm. The areas under the ROC curve (AUCs) of the random forest classifier (RFC) model, support vector machine (SVM), eXtreme gradient boosting (XGBoost), artificial neural network (ANN), and decision tree (DT) ranged from 0.725 (95% CI 0.223–1.227) to 0.902 (95% CI 0.400–1.403). Among them, RFC obtained the optimal prediction efficiency via adding inflammatory factors, which are serum creatinine (Scr), C-reactive protein (CRP), platelet-to-lymphocyte ratio (PLR), neutrophil-to-lymphocyte ratio (NLR), neutrophil-to-albumin ratio (NAR), and CysC, respectively. Conclusion We successfully developed ML-based prediction models for AKI, particularly the RFC, which can improve the prediction of AKI in patients with AP. The practicality of prediction and early detection may be greatly beneficial to risk stratification and management decisions.

Bibliometric and visual analysis of machine learning-based research in acute kidney injury worldwide

“…Among the top ten ranked studies, Zhang et al's study in 2019 included the smallest sample size, which still exceeded 6,000 cases (14), while Tomašev et al's study even exceeded 700,000 cases (21), which again proves the preference and applicability of machine learning for large sample size data and encourages future researchers to place more emphasis on sample size. However, it must be noted that the endpoint events of the current study are more focused on the AKI outcome of all-cause inpatients, while in studies of AKI prediction models related to specific comorbidities (23), specific nephrotoxic drugs (24), and specific procedures (25), the selection of sample size is influenced by morbidity, and given the scarcity of similar peer studies, the published literature of such studies is often underestimated. These highly cited studies also differed in terms of predictive endpoint events, model type, and predictive timeliness, with several studies using AKI onset as the predictive outcome and 48 h earlier as the timeliness assessment point (21,22), but some studies have developed a real-time prediction model considering the temporal changes in AKI events (26,27).…”

Predictive model of acute kidney injury in critically ill patients with acute pancreatitis: a machine learning approach using the MIMIC-IV database

“…Notable contributions in this domain include the work of Yi Yang et al. who developed machine learning-based prediction models tailored for acute AKI, emphasizing the potential of random forest classifiers to enhance predictive efficacy in patients with acute pancreatitis [ 22 ]. Similarly, Yang Fei et al.…”

Establishment and Validation of an Early Predictive Model for Severe Acute Pancreatitis

“…28 Acute kidney injury (AKI) is a common complication of SAP. 29 The changes in early SCR levels, especially within 24 hours after admission, are effective predictive indicators of the severity of AP. 30 Hypertriglyceridemia (HTG) is a risk factor for AP.…”