Machine Learning Prediction Model for Acute Renal Failure After Acute Aortic Syndrome Surgery

Abstract

BackgroundAcute renal failure (ARF) is the most common major complication following cardiac surgery for acute aortic syndrome (AAS) and worsens the postoperative prognosis. Our aim was to establish a machine learning prediction model for ARF occurrence in AAS patients.MethodsWe included AAS patient data from nine medical centers (n = 1,637) and analyzed the incidence of ARF and the risk factors for postoperative ARF. We used data from six medical centers to compare the performance of four machine learning models and performed internal validation to identify AAS patients who developed postoperative ARF. The area under the curve (AUC) of the receiver operating characteristic (ROC) curve was used to compare the performance of the predictive models. We compared the performance of the optimal machine learning prediction model with that of traditional prediction models. Data from three medical centers were used for external validation.ResultsThe eXtreme Gradient Boosting (XGBoost) algorithm performed best in the internal validation process (AUC = 0.82), which was better than both the logistic regression (LR) prediction model (AUC = 0.77, p < 0.001) and the traditional scoring systems. Upon external validation, the XGBoost prediction model (AUC =0.81) also performed better than both the LR prediction model (AUC = 0.75, p = 0.03) and the traditional scoring systems. We created an online application based on the XGBoost prediction model.ConclusionsWe have developed a machine learning model that has better predictive performance than traditional LR prediction models as well as other existing risk scoring systems for postoperative ARF. This model can be utilized to provide early warnings when high-risk patients are found, enabling clinicians to take prompt measures.

Comparing machine learning and logistic regression for acute kidney injury prediction in trauma patients: A retrospective observational study at a single tertiary medical center

“…[ 10 , 11 , 26 ] Although the XG Boost algorithm is known to exhibit better performance than other algorithms, it requires more time for model training. [ 25 , 27 ] Our study demonstrated a good classification performance with only hyperparameter tuning through Grid Search CV, without using XG Boost or gradient boosting tree algorithms.…”

“…The SHAP method intuitively shows the contribution of each feature and is effective in determining the importance of individual features. [ 11 , 25 ] The confusion matrix was categorized into 4 groups based on the classification outcome and is most commonly used for evaluating binary classifiers. [ 11 ] The F1 score is the harmonic mean of precision and recall.…”

The shared biomarkers and pathways of systemic lupus erythematosus and metabolic syndrome analyzed by bioinformatics combining machine learning algorithm and single-cell sequencing analysis

“…Therefore, we utilized machine learning to ascertain whether these 20 hub genes can construct a comprehensive prognostic model. XGBoost algorithm, a popular algorithm in machine learning classifiers that has demonstrated excellent performance ( 28 ), was selected for generating the model based on those 20 hub genes. In order to improve the model, XGBoost calculates negative gradients and uses them to find problems.…”

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

“…There were 1,909 articles identified [Cochrane ( n = 133), PubMed ( n = 33), Embase ( n = 231), Web of Science ( n = 1,512)]. After removing 220 duplicates, titles and abstracts of the remaining 1,689 articles were browsed, and 38 studies ( 12 – 49 ) were finally included. A PRISMA flow diagram of the study selection process is shown in Supplementary File 4 .…”

“…These were 60 prognostic models for CSA-AKI included, 12 external validation models, and 7 random sampling validation models. The types of these 60 prognostic models include: Logistic Regression ( 12 – 16 , 19 – 22 , 24 – 35 , 37 – 49 ) ( n = 34), Neural Net ( 15 , 17 – 19 ) ( n = 6), Support Vector Machine ( 15 , 16 ) ( n = 4), Random Forest ( 15 , 16 , 30 , 40 ) ( n = 6), Extreme Gradient Boosting ( 15 , 16 , 49 ) ( n = 3), Decision Tree ( 15 , 16 ) ( n = 3), Gradient Boosted Machine ( 19 ) ( n = 1), COX regression ( 19 ) ( n = 1), κ Neural Net ( 19 ) ( n = 1), and Naïve Bayes ( 19 ) ( n = 1). Characteristics of included studies are shown in Supplementary File 1 .…”