Skip to main content
. 2019 Apr 26;2:29. doi: 10.1038/s41746-019-0104-2

Fig. 2.

Fig. 2

Flowchart shows the summary of the data processing from bagging in the training phase to the final evaluation phase. Briefly, we obtained 10 ResNet models for predicting continuous eGFR and 10 XGBoost models for CKD status classification. In the evaluation phase, we averaged the output of 10 models as the final prediction result