Prediction of Visual Acuity after anti-VEGF Therapy in Diabetic Macular Edema by Machine Learning

Table 3

Accuracy of visual acuity variance predictions.

Algorithm learner	MAE		MSE
	All features	Selected features	All features	Selected features
Linear regression (LR)	0.188	0.188	0.069	0.069
SVM	0.254	0.349	0.111	0.208
K neighbors regressor	0.187	0.202	0.065	0.071
Random forest regressor (RF)	0.193	0.185	0.071	0.069
Ridge regressor	0.188	0.187	0.069	0.069
LR + RF	0.169	0.164	0.059	0.056

MAE, mean absolute error; MSE, mean square error; accuracy (VA in logMAR) of VA variance prediction at 1 month after anti-VEGF compared with the ground truth.