TY - JOUR
T1 - DenseUNets with feedback non-local attention for the segmentation of specular microscopy images of the corneal endothelium with guttae
AU - Vigueras-Guillén, Juan Pedro
AU - van Rooij, Jeroen
AU - van Dooren, Bart T H
AU - Lemij, Hans G
AU - Islamaj, Esma
AU - van Vliet, Lucas J
AU - Vermeer, Koenraad Arndt
N1 - © 2022. The Author(s).
PY - 2022/8/18
Y1 - 2022/8/18
N2 - Corneal guttae, which are the abnormal growth of extracellular matrix in the corneal endothelium, are observed in specular images as black droplets that occlude the endothelial cells. To estimate the corneal parameters (endothelial cell density [ECD], coefficient of variation [CV], and hexagonality [HEX]), we propose a new deep learning method that includes a novel attention mechanism (named fNLA), which helps to infer the cell edges in the occluded areas. The approach first derives the cell edges, then infers the well-detected cells, and finally employs a postprocessing method to fix mistakes. This results in a binary segmentation from which the corneal parameters are estimated. We analyzed 1203 images (500 contained guttae) obtained with a Topcon SP-1P microscope. To generate the ground truth, we performed manual segmentation in all images. Several networks were evaluated (UNet, ResUNeXt, DenseUNets, UNet++, etc.) and we found that DenseUNets with fNLA provided the lowest error: a mean absolute error of 23.16 [cells/mm2] in ECD, 1.28 [%] in CV, and 3.13 [%] in HEX. Compared with Topcon’s built-in software, our error was 3–6 times smaller. Overall, our approach handled notably well the cells affected by guttae, detecting cell edges partially occluded by small guttae and discarding large areas covered by extensive guttae.
AB - Corneal guttae, which are the abnormal growth of extracellular matrix in the corneal endothelium, are observed in specular images as black droplets that occlude the endothelial cells. To estimate the corneal parameters (endothelial cell density [ECD], coefficient of variation [CV], and hexagonality [HEX]), we propose a new deep learning method that includes a novel attention mechanism (named fNLA), which helps to infer the cell edges in the occluded areas. The approach first derives the cell edges, then infers the well-detected cells, and finally employs a postprocessing method to fix mistakes. This results in a binary segmentation from which the corneal parameters are estimated. We analyzed 1203 images (500 contained guttae) obtained with a Topcon SP-1P microscope. To generate the ground truth, we performed manual segmentation in all images. Several networks were evaluated (UNet, ResUNeXt, DenseUNets, UNet++, etc.) and we found that DenseUNets with fNLA provided the lowest error: a mean absolute error of 23.16 [cells/mm2] in ECD, 1.28 [%] in CV, and 3.13 [%] in HEX. Compared with Topcon’s built-in software, our error was 3–6 times smaller. Overall, our approach handled notably well the cells affected by guttae, detecting cell edges partially occluded by small guttae and discarding large areas covered by extensive guttae.
KW - Cell Count
KW - Endothelial Cells
KW - Endothelium, Corneal/diagnostic imaging
KW - Feedback
KW - Microscopy/methods
UR - https://www.mendeley.com/catalogue/cbdf7d09-d034-3e20-a403-b28c46938d8a/
U2 - 10.1038/s41598-022-18180-1
DO - 10.1038/s41598-022-18180-1
M3 - Article
C2 - 35982194
SN - 2045-2322
VL - 12
SP - 14035
JO - Scientific Reports
JF - Scientific Reports
IS - 1
ER -