Key result
Extreme gradient boosting outperforms logistic regression for predicting post-cardiac surgery AKI with ~0.78 AUC.
Why the study?
Do machine learning approaches improve the prediction of acute kidney injury after cardiac surgery compared to logistic regression analysis?
Population
2010 patients who underwent open heart surgery and thoracic aortic surgery
Comparison
Machine learning techniques for risk prediction vs Logistic regression analysis
Design
Cohort
Follow-up
1 week
Authors
Loading...
May support ML for postoperative AKI stratification; hypothesis-generating and requires prospective validation before practice change.
Observational (n=2,010)
No
Do machine learning approaches improve the prediction of acute kidney injury after cardiac surgery compared to logistic regression analysis?
Effect estimate: AUC 0.78 (95% CI 0.75-0.80)
Absolute Event Rate: 0.78% vs 0.69%
p-value: p=<0.001
Gradient boosting machine learning techniques outperform traditional logistic regression in predicting acute kidney injury after cardiac surgery.
Lee et al. (2018) conducted an observational in Acute kidney injury after cardiac surgery (n=2,010). Extreme gradient boosting machine learning vs. Logistic regression analysis was evaluated on Area under the receiver-operating characteristic curve (AUC) for predicting postoperative acute kidney injury (AKI) (AUC 0.78, 95% CI 0.75-0.80, p=<0.001). Extreme gradient boosting machine learning achieved a significantly higher area under the curve (0.78) compared to logistic regression (0.69) for predicting acute kidney injury after cardiac surgery.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: