AUTHOR=Song Zhe , Yang Zhenyu , Hou Ming , Shi Xuedong 

TITLE=Machine learning in predicting cardiac surgery-associated acute kidney injury: A systemic review and meta-analysis

JOURNAL=Frontiers in Cardiovascular Medicine

VOLUME=Volume 9 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/cardiovascular-medicine/articles/10.3389/fcvm.2022.951881

DOI=10.3389/fcvm.2022.951881

ISSN=2297-055X

ABSTRACT=Background Cardiac surgery-associated acute kidney injury (CSA-AKI) is a common complication following cardiac surgery. Early prediction of CSA-AKI is of great significance for improving patients` prognosis. The aim of this study is to systematically evaluate the predictive performance of machine learning models for CSA-AKI.
Methods Cochrane Library, PubMed, EMBASE, and Web of Science were searched from inception to March 18, 2022. Risk of bias assessment was performed using PROBAST. R software (version 4.1.1) was used to calculate the accuracy and C-index of CSA-AKI predicting. The importance of CSA-AKI predicting was defined according to frequency of related factors in the models. 
Results There were 38 eligible studies included, with a total of 255943 patients and 60 machine learning models. The models mainly included Logistic Regression (n=34), Neural Net (n=6), Support Vector Machine (n=4), Random Forest (n=6), Extreme Gradient Boosting (n=3), Decision Tree (n=3), Gradient Boosted Machine (n=1), COX regression (n=1),κNeural Net (n=1), and Naïve Bayes (n=1), of which 51 models with intact recording in training set and 17 in validating set. Variables with highest predicting frequency included Logistic Regression、Neural Net、Support Vector Machine、Random Forest. The C-index and accuracy were 0.76 (0.740, 0.780) and 0.72 (0.70, 0.73), respectively, in training set, and 0.79 (0.75, 0.83) and 0.73 (0.71,0.74), respectively, in test set.
Conclusion Machine learning-based model is effective for the early prediction of CSA-AKI. More machine learning methods based on non-invasive or minimally invasive predictive indicators are needed to improve the predictive performance and make accurate prediction of CSA-AKI. Logistic regression remains currently the most commonly applied model in CSA-AKI prediction, though it is not the one with the best performance. There are other models that would be more effective, such as NNET and XGBoost.