Prediction of Acute Kidney Injury in Intracerebral Hemorrhage Patients Using Machine Learning

Abstract

Background: Acute kidney injury (AKI) is prevalent in patients with intracerebral hemorrhage (ICH) and is associated with mortality. This study aimed to verify the predictive accuracy of different machine learning algorithms for AKI in patients with ICH using a large dataset. Methods: A total of 1366 ICH patients received treatments between 2001 and 2012 from the Medical Information Mart for Intensive Care-III (MIMIC-III) database were identified based on the ICD-9 code: 431. The main outcome of AKI during hospitalizations was confirmed based on the KDIGO criteria. Overall, ICH patients were randomly divided into the training cohort and validation cohort with the ratio of 7:3. Six machine learning algorithms including extreme gradient boosting, logistic, light gradient boosting machine, random forest, adaptive boosting, support vector machine were trained in the training cohort with the 5-fold cross-validation method to predict the AKI. The predictive accuracy of those algorithms was compared by area under the receiver operating characteristics curve (AUC). Results: A total of 1213 ICH patients were included with the incidence of AKI being 29.3%. The incidence of AKI was 29.3% among the 1213 patients with ICH. The AKI group had higher 30-day mortality (p<0.001), longer ICU stay (p<0.001), and longer hospital stay (p<0.001). Among the six machine learning algorithms, the random forest performed the best in predicting AKI in both the training cohort (AUC=1.000) and the validation cohort (AUC=0.698). The top five features in the random forest algorithm-based model were platelets, serum creatinine, vancomycin, hemoglobin, and hematocrit. Conclusion:The random forest algorithm-based predictive model we developed incorporating important features, including platelet count, serum creatinine level, vancomycin level, hemoglobin level, and hematocrit level, performed the best in predicting AKI among patients with ICH.

Acute kidney injury after intracerebral hemorrhage: a mini review

“…Among these, antibiotics such as vancomycin and diuretics are frequently implicated. She et al utilized the random forest algorithm and found that vancomycin emerged as a significant predictive factor for AKI in patients with ICH ( 35 ). Additionally, another study noted that diuretic escalated the risk of AKI in individuals with acute stroke (OR = 8.5, 95%CI: 3.46–20.86, p < 0.001) ( 26 ).…”

“…Qureshi’s research indicated that patients with ICH who had a baseline blood creatinine level ≥ 110 μmol/L faced an elevated risk of AKI and renal adverse events, with OR of 2.4 (95%CI, 1.2–4.5) and 3.1 (95%CI, 1.2–8.1), respectively ( 4 ). A study involving 1,366 ICH patients similarly highlighted creatinine as a characteristic of AKI ( 35 ). However, some studies have found no correlation between renal function and AKI, and the presence of CKD alongside AKI did not correlate with long-term mortality (OR = 3.13, 95%CI: 0.65–15.01, p = 0.154) ( 43 ).…”

“…In addition, She et al used six machine learning algorithms—extreme gradient boosting, logistic, light gradient boosting machine, random forest, adaptive boosting, and support vector machine—to develop the model. They found that the random forest algorithm performed the best in both the training and validation sets, achieving an AUC of 1.000 in the training set and 0.698 in the validation set ( 35 ). However, these models are constrained by relatively small sample sizes, limiting their credibility.…”