Identification of the Optimal Model for the Prediction of Diabetic Retinopathy in Chinese Rural Population: Handan Eye Study

Table 4

Performance of validation of the DR prediction models.

Model	Accuracy (95% CI)	Sensitivity	Specificity
GBM	0.883 (0.866, 0.900)	0.943	0.881
RF	0.843 (0.823, 0.862)	0.971	0.840
MLR	0.898 (0.881, 0.913)	0.914	0.898
SVMs	0.919 (0.966, 0.983)	0.571	0.928
C&RT	0.978 (0.968, 0.985)	0.371	0.992

Note: Gradient boosting machine (GBM); random forests (RF); multivariable logistic regression (MLR); support vector machine (SVM); classification and regression trees (C&RT).