Skip to main content

Table 2 Comparison of prediction accuracies for stone-free and one-session success according to three machine learning methods

From: Machine learning prediction of stone-free success in patients with urinary stone after treatment of shock wave lithotripsy

 

Stone-free

One-session success

Random forest (RF)

 Training Accuracy (%)

86.47

76.83

 Test Accuracy (%)

85.98

78.02

Extreme gradient boosting trees (XGBoost)

 Training Accuracy (%)

87.50

75.60

 Test Accuracy (%)

87.46

77.39

Light Gradient Boosting Method (LightGBM)

 Training Accuracy (%)

88.09

74.92

 Test Accuracy (%)

87.95

77.04