Key result
Preoperative automated machine learning outperforms standard logistic regression for predicting post-cardiac surgery acute kidney injury.
Why the study?
The authors sought to develop and validate an automated machine learning prediction model for cardiac surgery-associated acute kidney injury.
Does an autoML prediction model accurately predict post-operative acute kidney injury in adult patients undergoing cardiac surgery compared to traditional models?
Observational (n=13,158)
No
Does an autoML prediction model accurately predict post-operative acute kidney injury in adult patients undergoing cardiac surgery compared to traditional models?
An explainable preoperative autoML model demonstrated high predictive performance for cardiac surgery-associated acute kidney injury, outperforming traditional logistic regression.
May aid preoperative CSA-AKI risk stratification; leaves open prospective validation and outcome impact before clinical adoption.
BACKGROUND: We aimed to develop and validate an automated machine learning (autoML) prediction model for cardiac surgery-associated acute kidney injury (CSA-AKI). METHODS: Using 69 preoperative variables, we developed several models to predict post-operative AKI in adult patients undergoing cardiac surgery. Models included autoML and non-autoML types, including decision tree (DT), random forest (RF), extreme gradient boosting (XGBoost), and artificial neural network (ANN), as well as a logistic regression prediction model. We then compared model performance using area under the receiver operating characteristic curve (AUROC) and assessed model calibration using Brier score on the independent testing dataset. RESULTS: The incidence of CSA-AKI was 36%. Stacked ensemble autoML had the highest predictive performance among autoML models, and was chosen for comparison with other non-autoML and multivariable logistic regression models. The autoML had the highest AUROC (0.79), followed by RF (0.78), XGBoost (0.77), multivariable logistic regression (0.77), ANN (0.75), and DT (0.64). The autoML had comparable AUROC with RF and outperformed the other models. The autoML was well-calibrated. The Brier score for autoML, RF, DT, XGBoost, ANN, and multivariable logistic regression was 0.18, 0.18, 0.21, 0.19, 0.19, and 0.18, respectively. We applied SHAP and LIME algorithms to our autoML prediction model to extract an explanation of the variables that drive patient-specific predictions of CSA-AKI. CONCLUSION: We were able to present a preoperative autoML prediction model for CSA-AKI that provided high predictive performance that was comparable to RF and superior to other ML and multivariable logistic regression models. The novel approaches of the proposed explainable preoperative autoML prediction model for CSA-AKI may guide clinicians in advancing individualized medicine plans for patients under cardiac surgery.
No takes yet. Share an insight, caveat, or question.
Thongprayoon et al. (2022) conducted an observational in Cardiac surgery-associated acute kidney injury (n=13,158). Automated machine learning (autoML) prediction model vs. Multivariable logistic regression and non-automated ML models was evaluated on Prediction of post-operative acute kidney injury (AUROC) (AUROC 0.79, 95% CI 0.77-0.81, p=0.01). A preoperative automated machine learning prediction model achieved high predictive performance for cardiac surgery-associated acute kidney injury (AUROC 0.79), significantly outperforming traditional multivariable logistic regression.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: