# Machine learning prediction of moderate‐to‐severe acute kidney injury after ICU admission and cardiac surgery with urine trace elements

## TL;DR

This study uses machine learning and urine trace elements to predict moderate-to-severe kidney injury in ICU and cardiac surgery patients, showing promising early detection potential.

## Contribution

The study introduces a novel application of urinary trace elements in machine learning models for early prediction of acute kidney injury.

## Key findings

- LightGBM, Random Forest, and XGBoost showed the highest predictive performance with AUCs between .71 and .85.
- Age, strontium, and boron were consistently top features across multiple models.
- ML models using urinary trace elements can help identify AKI risk in ICU and cardiac surgery patients.

## Abstract

Acute kidney injury (AKI) is common and linked to poor outcomes, but early detection remains challenging. Previous research identified urinary trace elements (TE) as early AKI biomarkers in intensive care unit (ICU) or cardiac surgery patients. We aimed to explore whether urinary TE enhance machine learning (ML) models for AKI prediction.

We constructed ML models using the ICU cohort. We filtered the variables and optimized hyperparameters before predicting Kidney Disease: Improving Global Outcomes stage 2–3 AKI using eight ML classifiers: light gradient boosting machine (LightGBM), random forest (RF), ML logistic regression, support vector machine, multilayer perceptron, eXtreme gradient boosting (XGBoost), Gaussian Naive Bayes and k‐nearest neighbors. External validation was performed in the cardiac surgery cohort.

Among 149 ICU patients (median age 56.0 [interquartile range (IQR): 43.5–67.0], 63.1% male), 25 developed stage 2–3 AKI; among 144 cardiac surgery patients (median age 70.0 [IQR: 62.0–76.0], 72.9% male), 12 developed stage 2–3 AKI. Each ML in the internal validation had area under the curve (AUC) above .7, with XGBoost having the highest (.813); LightGBM had the second highest AUC (.799), highest G‐mean (.567) and F1‐score (.545). In external validation, RF had the highest AUC (.740), XGBoost had the highest G‐mean (.289) and F1‐score (.286). Age, strontium and boron were consistently ranked among the top five most important features in LightGBM, RF and XGBoost.

ML models primarily based on urinary TE can identify AKI risk in both clinical groups (ICU and cardiac surgery), with LightGBM, RF and XGBoost serving as high‐performance models for early prediction of stage 2–3 AKI.

Acute kidney injury (AKI) is a frequent complication after ICU admission and cardiac surgery. In this study, 149 ICU patients and 144 post–cardiac surgery patients were evaluated using urinary trace elements with eight machine learning (ML) algorithms. LightGBM, Random Forest and XGBoost showed the highest predictive performance (AUCs .71–.85). Key predictive features included age, vanadium, boron, copper, molybdenum, strontium, albumin, zinc, phosphorus and lithium. ML models incorporating urinary trace elements may enable early stratification of moderate‐to‐severe AKI risk.