Prediction of premature all-cause mortality in patients receiving peritoneal dialysis using modified artificial neural networks

Qiongxiu Zhou; Xiaohan You; Haiyan Dong; Zhe Lin; Yanling Shi; Zhen Su; Rongrong Shao; Chaosheng Chen; Ji Zhang

doi:doi:10.18632/aging.203033

← Back

Research Paper Volume 13, Issue 10 pp 14170—14184

Prediction of premature all-cause mortality in patients receiving peritoneal dialysis using modified artificial neural networks

Figure 4. Permutation feature importance for the ANN models in the total dataset (0-month), 6-month dataset and 12-month dataset. Higher positive values indicated greater importance of the model, and negative values may indicate that the feature is worse than noise.