Deep Learning as a Tool for Automatic Segmentation of Corneal Endothelium Images

The automatic analysis of the state of the corneal endothelium is of much interest in ophthalmology. Up till now, several manual and semi-automatic methods have been introduced, but the need of fully-automatic segmentation of cells in the endothelium is still in search. This work addresses the problem of automatic delineation of cells in the corneal endothelium images and suggests to use the convolutional neural network (CNN) to classify between cell center, cell body, and cell border in order to achieve precise segmentation. Additionally, a method to automatically select and split merged cells is given. In order to skeletonize the result, the best-fit method is used. The achieved outcomes are compared to manual annotations in order to define the mutual overlapping. The Dice index, Jaccard coefficient, modified Hausdorff distance, and several other metrics for mosaic overlapping are used. As a final check-up, the visual inspection is shown. The performed experiments revealed the best architecture for CNN. The correctness and precision of the segmentation were evaluated on Endothelial Cell “Alizarine” dataset. According to the Dice index and Jaccard coefficient, the automatically achieved cell delineation overlaps the original one with 93% precision. While modified Hausdorff distance shows 0.14 pixel distance, proving very high accuracy. These findings are confirmed by other metrics and also supported by presented visual inspection of achieved segmentations. To conclude, the methodology to achieve fully-automatic delineation of cell boundaries in the corneal endothelium images was presented. The segmentation obtained as a result of pixel classification with CNN proved very high precision.

[1]  Cris L. Luengo Hendriks,et al.  Improving the stochastic watershed , 2013, Pattern Recognit. Lett..

[2]  Motohide Yoshimura,et al.  Convolution neural network for contour extraction of corneal endothelial cells , 2017, International Conference on Quality Control by Artificial Vision.

[3]  Karolina Nurzynska,et al.  Influence of applied corneal endothelium image segmentation techniques on the clinical parameters , 2017, Comput. Medical Imaging Graph..

[4]  A. Ruggeri,et al.  A system for the automatic estimation of morphometric parameters of corneal endothelium in alizarine red-stained images , 2010, British Journal of Ophthalmology.

[5]  Adam Piórkowski,et al.  Cell Detection in Corneal Endothelial Images Using Directional Filters , 2015, IP&C.

[6]  Anna Fabijanska,et al.  Corneal endothelium image segmentation using feedforward neural network , 2017, 2017 Federated Conference on Computer Science and Information Systems (FedCSIS).

[7]  M. Doughty The ambiguous coef f icient of variation: Polymegethism of the corneal endothelium and central corneal thickness , 1990 .

[8]  S. Duke-Elder,et al.  Textbook of Ophthalmology , 1950 .

[9]  Filip Malmberg,et al.  Exact Evaluation of Stochastic Watersheds: From Trees to General Graphs , 2014, DGCI.

[10]  Dominik Sankowski,et al.  Noise adaptive switching median-based filter for impulse noise removal from extremely corrupted images , 2011 .

[11]  Marek Kretowski,et al.  MESA: Complete approach for design and evaluation of segmentation methods using real and simulated tomographic images , 2014 .

[12]  Roberto de Alencar Lotufo,et al.  Identification and segmentation of cells in images of donated corneas using mathematical morphology , 2000, Proceedings 13th Brazilian Symposium on Computer Graphics and Image Processing (Cat. No.PR00878).

[13]  Anil K. Jain,et al.  A modified Hausdorff distance for object matching , 1994, Proceedings of 12th International Conference on Pattern Recognition.

[14]  M. Foracchia,et al.  Corneal endothelium analysis by means of Bayesian shape modeling , 2003, Proceedings of the 25th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (IEEE Cat. No.03CH37439).

[15]  M. Aurangzeb Khan,et al.  Endothelial Cell Image Enhancement using Non-subsampled Image Pyramid , 2007 .

[16]  Karolina Nurzynska,et al.  Shape parameters for automatic classification of snow particles into snowflake and graupel , 2013 .

[17]  Karolina Nurzynska,et al.  THE CORRELATION ANALYSIS OF THE SHAPE PARAMETERS FOR ENDOTHELIAL IMAGE CHARACTERISATION , 2016 .

[18]  J. Aquavella,et al.  Cell size-shape relationships in corneal endothelium. , 1982, Investigative ophthalmology & visual science.

[19]  Issam Dagher,et al.  WaterBalloons: A Hybrid Watershed Balloon Snake Segmentation , 2007, 2007 International Joint Conference on Neural Networks.

[20]  H. Edelhauser,et al.  Corneal endothelial morphology in the rat. Effects of aging, diabetes, and topical aldose reductase inhibitor treatment. , 1988, Investigative ophthalmology & visual science.

[21]  Adam Piórkowski,et al.  Towards Automated Cell Segmentation in Corneal Endothelium Images , 2014, IP&C.

[22]  Nigel H Brookes,et al.  Morphometry of organ cultured corneal endothelium using Voronoi segmentation , 2017, Cell and Tissue Banking.

[23]  J. Pinoli,et al.  Human visual perception and dissimilarity , 2013 .

[24]  Adam Piórkowski Best-Fit Segmentation Created Using Flood-Based Iterative Thinning , 2016, IP&C.

[25]  M. Doughty Concerning the symmetry of the 'hexagonal' cells of the corneal endothelium. , 1992, Experimental eye research.

[26]  Khalid Saeed,et al.  K3M: A universal algorithm for image skeletonization and a review of thinning techniques , 2010, Int. J. Appl. Math. Comput. Sci..

[27]  Przemyslaw Mazurek,et al.  Cell structures modeling using fractal generator and torus geometry , 2016, 2016 21st International Conference on Methods and Models in Automation and Robotics (MMAR).

[28]  A. Ruggeri,et al.  Corneal Endothelium Cell Field Analysis by means of Interacting Bayesian Shape Models , 2007, 2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[29]  Luc Vincent,et al.  Morphological image processing and network analysis of cornea endothelial cell images , 1992, Optics & Photonics.

[30]  Filip Malmberg,et al.  Fast Evaluation of the Robust Stochastic Watershed , 2015, ISMM.

[31]  Kazuyoshi Itoh,et al.  Generalization of shift invariant neural networks: Image processing of corneal endothelium , 1996, Neural Networks.

[32]  L. R. Dice Measures of the Amount of Ecologic Association Between Species , 1945 .

[33]  Cris L. Luengo Hendriks,et al.  Fully automatic evaluation of the corneal endothelium from in vivo confocal microscopy , 2015, BMC Medical Imaging.

[34]  D. Reska,et al.  Automatic segmentation of corneal endothelial cells using active contours , 2014 .

[35]  M. Foracchia,et al.  Cell contour detection in corneal endothelium in-vivo microscopy , 2000, Proceedings of the 22nd Annual International Conference of the IEEE Engineering in Medicine and Biology Society (Cat. No.00CH37143).

[36]  Jean-Charles Pinoli,et al.  A Geometric Dissimilarity Criterion Between Jordan Spatial Mosaics. Theoretical Aspects and Application to Segmentation Evaluation , 2011, Journal of Mathematical Imaging and Vision.

[37]  B. Heise,et al.  CELL SEGMENTATION USING LEVEL SET METHOD , 2007 .