DenseUNets with feedback non-local attention for the segmentation of specular microscopy images of the corneal endothelium with Fuchs dystrophy
To estimate the corneal endothelial parameters from specular microscopy images depicting cornea guttata (Fuchs endothelial dystrophy), we propose a new deep learning methodology that includes a novel attention mechanism named feedback non-local attention (fNLA). Our approach first infers the cell edges, then selects the cells that are well detected, and finally applies a postprocessing method to correct mistakes and provide the binary segmentation from which the corneal parameters are estimated (cell density [ECD], coefficient of variation [CV], and hexagonality [HEX]). In this study, we analyzed 1203 images acquired with a Topcon SP-1P microscope, 500 of which contained guttae. Manual segmentation was performed in all images. We compared the results of different networks (UNet, ResUNeXt, DenseUNets, UNet++) and found that DenseUNets with fNLA provided the best performance, with a mean absolute error of 23.16 [cells/mm^2] in ECD, 1.28 [ in HEX, which was 3-6 times smaller than the error obtained by Topcon's built-in software. Our approach handled the cells affected by guttae remarkably well, detecting cell edges occluded by small guttae while discarding areas covered by large guttae. fNLA made use of the local information, providing sharper edges in guttae areas and better results in the selection of well-detected cells. Overall, the proposed method obtained reliable and accurate estimations in extremely challenging specular images with guttae, being the first method in the literature to solve this problem adequately. Code is available in our GitHub.
READ FULL TEXT