Correction: A deep learning model for the detection of both advanced and early glaucoma using fundus photography( PLoS ONE (2019) 14:1 (e0211579) DOI: 10.1371/journal.pone.0211579)

Abstract

There are errors in the second and third sentence of the second paragraph under the subheading “Training Model” in the Methods section. The correct sentences are: Two convolutional layers, with patch sizes of 20x20 and 40x40, were used with a stride of 1 and depths of 16 and 32. Max pooling was applied, with a patch size of 2x2 and a stride of 2. Fig 3 is incorrect. The text under both “Max-pooling” labels should read “2x2 kernel.” The authors have provided a corrected version here.