Fig. 1

Schematic diagram of the overall analysis workflow. Candidate features were first grouped into preoperative, intraoperative, and postoperative feature sets. Preoperative and intraoperative features were merged into a set of \({F}_{pre}\), which were used in all the tested models. Postoperative features were categorized based on the sampling time points (postoperative day [POD] 0, 1, 2, 3, 4, 5, and postoperative 1 month). Each of these feature sets was combined with \({F}_{pre}\) to yield 7 different feature sets (\({F}_{0d}\), …, \({F}_{5d}\), and \({F}_{1m}\)). The 8 feature sets were used to fit 8 different Lasso regression models. Features of each set with non-zero coefficients, \({F}_{Lasso}\), were then passed onto a partial correlation filter that evaluated the correlation of each of the features with the target variable \(\Delta {SCr}_{1y}\). The final features, \({F}_{final}\), were then used to train the final multivariate linear regression models. The final step used all possible combinations of the predictions generated by the 7 final models, \({Model}_{0d-1m}\), to yield 120 meta-models