Erratum: Machine learning-based prediction of off-pump coronary artery bypass grafting-associated acute kidney injury

In the July 30, 2024 issue of J Thorac Dis, the paper “Machine learning-based prediction of off-pump coronary artery bypass grafting-associated acute kidney injury” by Dr. Song et al. (1) was published with some errors.

Corrections are shown below:

(I) In the section of Results, the sentences “Table 2 shows the performance of these models when intraoperative data were added to the baseline data; the AUC of the GBDT model was again the highest (AUC =0.861), followed by RF model (AUC =0.780), XGBoost model (AUC =0.764), SVM model (AUC =0.730), AdaBoost model (AUC =0.726), LR model (AUC =0.700), KNN model (AUC =0.598) and DT model (AUC =0.550). These data demonstrate that the addition of intraoperative time series data resulted in a considerable increase in AUC; in case of the GBDT model, AUC increased by 0.122.” should be corrected to “Table 2 shows the performance of these models when intraoperative data were added to the baseline data; the AUC of the GBDT model was again the highest (AUC =0.835), followed by SVM model (AUC =0.737), LR model (AUC =0.709), XGBoost model (AUC =0.703), RF model (AUC =0.656), AdaBoost model (AUC =0.578), KNN model (AUC =0.573) and DT model (AUC =0.443). These data demonstrate that the addition of intraoperative time series data resulted in a considerable increase in AUC; in case of the GBDT model, AUC increased by 0.096.”

The corrected Table 2 appears below.

Table 2. AUC and accuracy rate of eight machine learning models based on baseline and intraoperative datasets.

Machine learning model	AUC	Accuracy rate
LR	0.709	0.786
SVM	0.737	0.829
DT	0.443	0.829
RF	0.656	0.936
KNN	0.573	0.936
GBDT	0.835	0.929
AdaBoost	0.578	0.907
XGBoost	0.703	0.943

AUC, area under the curve; LR, logistic regression; SVM, support vector machine; DT, decision tree; RF, random forest; KNN, k-nearest neighbor; GBDT, gradient-boosting decision tree; AdaBoost, adaptive boosting; XGBoost, eXtreme gradient boosting.

Accordingly, the results presented in the abstract section require modification. Specifically, the intraoperative datasets should be updated to an AUC of 0.835 and an accuracy of 0.929, instead of the previously reported AUC of 0.861 and accuracy of 0.936.

(II) In the section of “Discussion”, “LS-LSTM” should be corrected as “CONV-LSTM”.

(III) In the “Abstract” section and the “Statistical analysis” section, “a long short-term memory (LSTM) deep learning model” should be changed to “a long short-term memory (CONV-LSTM) deep learning model”.

The authors regret the errors and confirm they will not change the conclusions of the article.

Click here to view the updated version of the article.

Supplementary

The article’s supplementary files as