AUTHOR=Shi Xi , Qu Tingyu , Van Pottelbergh Gijs , van den Akker Marjan , De Moor Bart 

TITLE=A Resampling Method to Improve the Prognostic Model of End-Stage Kidney Disease: A Better Strategy for Imbalanced Data

JOURNAL=Frontiers in Medicine

VOLUME=Volume 9 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/medicine/articles/10.3389/fmed.2022.730748

DOI=10.3389/fmed.2022.730748

ISSN=2296-858X

ABSTRACT=Background: Prognostic models can help to identify patients at risk for End-Stage-Kidney Disease (ESKD) at an earlier stage to provide preventive medical interventions. Previous studies mostly applied Cox proportional hazards model. The aim of this study is to present a resampling method, which can deal with imbalanced data structure for the prognostic model and help improve predictive performance. 
Methods: The Electronic Health Records of Chronic Kidney Disease (CKD) patients older than 50 during 2005-2015 collected from primary care in Belgium were used (n=11,645). Both Cox model and logistic regression were applied as reference model. Then the resampling method, SMOTE-ENN, was applied as a preprocessing procedure followed by the logistic regression. The performance was evaluated by accuracy, AUC, confusion matrix, and F3 score. 
Results: The C statistics for Cox model was 0.807 while the AUC for logistic regression was 0.700, both on a comparable level to previous studies. With the model trained on the resampled set, 86.3% of ESKD patients were correctly identified, although it was at the cost of the high misclassification rate of negative cases. The F3 score was 0.245, much higher than 0.043 for logistic regression and 0.022 for Cox model. 
Conclusions: Our study pointed out the imbalanced data structure and its effects on prediction accuracy, which were not thoroughly discussed in previous studies. We were able to identify patients with high risk for ESKD better from a clinical perspective by using the resampling method. But it has the limitation of the high misclassification of negative cases. The technique can be widely used in other clinical topics when imbalanced data structure should be considered.