Abstract
Purpose
Specular microscopy is an indispensable tool for clinicians seeking to monitor the corneal endothelium. Automated methods of determining endothelial cell density (ECD) are limited in their ability to analyze images of poor quality. We describe and assess an image processing algorithm to analyze corneal endothelial images.
Methods
A set of corneal endothelial images acquired with a Konan CellChek specular microscope was analyzed using three methods: flex-center, Konan Auto Tracer, and the proposed method. In this technique, the algorithm determines the region of interest, filters the image to differentiate cell boundaries from their interiors, and utilizes stochastic watershed segmentation to draw cell boundaries and assess ECD based on the masked region. We compared ECD measured by the algorithm with manual and automated results from the specular microscope.
Results
We analyzed a total of 303 images manually, using the Auto Tracer, and with the proposed image processing method. Relative to manual analysis across all images, the mean error was 0.04% in the proposed method (p = 0.23 for difference) whereas Auto Tracer demonstrated a bias towards overestimation, with a mean error of 5.7% (p = 2.06× 10-8). The relative mean absolute errors were 6.9% and 7.9%, respectively, for the proposed and Auto Tracer. The average time for analysis of each image using the proposed method was 2.5 s.
Conclusion
We demonstrate a computationally efficient algorithm to analyze corneal endothelial cell density that can be implemented on devices for clinical and research use.
Similar content being viewed by others
Data availability
The specific images used in the research cannot be shared in public due to the patient consent; however, the images were captured using Konan specular microscope and authors believe the algorithm should work with similar accuracy with any images taken using the similar device.
Code availability
Authors implemented their own code on MATLAB.
Change history
29 November 2021
A Correction to this paper has been published: https://doi.org/10.1007/s00417-021-05512-6
References
McCarey BE (1979) Noncontact specular microscopy: a macrophotography technique and some endothelial cell findings. Ophthalmology 86(10):1848–1860
Bourne WM, Nelson LR, Hodge DO (1997) Central corneal endothelial cell changes over a ten-year period. Invest Ophthalmol Vis Sci 38(3):779–782
Laing RA, Sandstrom MM, Berrospi AR et al (1976) Changes in the corneal endothelium as a function of age. Exp Eye Res 22(6):587–594
Mohammad-Salih P (2011) Corneal endothelial cell density and morphology in normal Malay eyes. Med J Malaysia 66(4):300–303
Rao SK, Sen PR, Fogla R et al (000) Corneal endothelial cell density and morphology in normal Indian eyes. Cornea 19(6):820–823
Maurice D (1968) Cellular membrane activity in the corneal endothelium of the intact eye. Cell Mol Life Sci 24(11):1094–1095
Laing RA, Sandstrom MM, Leibowitz HM (1975) Vivo photomicrography of the corneal endothelium. Arch Ophthalmol 93(2):143–145
Bourne WM, Kaufman HE (1976) Specular microscopy of human corneal endothelium in vivo. Am J Ophthalmol 81(3):319–323
Jalbert I, Stapleton F, Papas E et al (2003) In vivo confocal microscopy of the human cornea. Br J Ophthalmol 87(2):225–236
Huang J, Maram J, Tepelus TC et al (2018) Comparison of noncontact specular and confocal microscopy for evaluation of corneal endothelium. Eye Contact Lens 44:144–150
Huang J, Maram J, Modak C et al (2014) Comparison of non-contact specular and confocal microscopy for the evaluation of the corneal endothelium. Invest Ophthalmol Vis Sci 55(13):999–999
Hara M, Morishige N, Chikama T et al (2003) Comparison of confocal biomicroscopy and noncontact specular microscopy for evaluation of the corneal endothelium. Cornea 22(6):512–515
Price MO, Fairchild KM, Price FW Jr (2013) Comparison of manual and automated endothelial cell density analysis in normal eyes and dsek eyes. Cornea 32(5):567–573
Miyagi H, Stanley AA, Chokshi TJ et al (2020) Comparison of automated vs manual analysis of corneal endothelial cell density and morphology in normal and corneal endothelial dystrophy-affected dogs. Vet Ophthalmol 23(1):44–51
Huang J, Maram J, Tepelus TC et al (2018) Comparison of manual & automated analysis methods for corneal endothelial cell density measurements by specular microscopy. J Opt 11(3):182–191
Villalba R, Jimenez A, Fornes G et al (2014) Flex center method versus center method for endothelial corneal evaluation in eye banking. a comparative analysis. Cell Tissue Bank 15(4):507–512
Thuret G, Manissolle C, Acquart S et al (2003) Is manual counting of corneal endothelial cell density in eye banks still acceptable? the french experience. Br J Ophthalmol 87(12):1481–1486
Vincent LM, Masters BR (1992) Morphological image processing and network analysis of cornea endothelial cell images. In: Image Algebra and Morphological Image Processing III, vol. 1769, pp. 212–226. International Society for Optics and Photonics
Salerno M, Sargeni F, Bonaiuto V, etal (1998) A new CNN based tool for an automated morphometry analysis of the corneal endothelium. In: Proceedings of Fifth IEEE International Workshop on Cellular Neural Networks and Their Applications. (Cat. No. 98TH8359), pp. 169–174
Foracchia M, Ruggeri A, etal. (2002), Estimating cell density in corneal endothelium by means of Fourier analysis. In: Proceedings of the Second Joint 24th Annual Conference and the Annual Fall Meeting of the Biomedical Engineering Society, Engineering in Medicine and Biology, vol. 2, pp. 1097–1098.
Grisan E, Paviotti A, Laurenti N, etal (2006), A lattice estimation approach for the automatic evaluation of corneal endothelium density. In: 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference, pp. 1700–1703.
Foracchia M, Ruggeri A (2007) Corneal endothelium cell field analysis by means of interacting bayesian shape models. In: 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, pp. 6035–6038.
Ruggeri A, Scarpa F, De Luca M et al (2012) A system for the automatic estimation of morphometric parameters of corneal endothelium in alizarine red-stained images. Br J Ophthalmol 94(5):643–647
Kumar KK, Srinivasa G (2018) Corneal endothelium cell segmentation and count using k-means and watershed algorithms. In: Second International Conference on Advances in Electronics, Computers and Communications (ICAECC), pp. 1–7. IEEE
Selig B, Vermeer KA, Rieger B et al (2015) Fully automatic evaluation of the corneal endothelium from in vivo confocal microscopy. BMC Med Imaging 15(1):13
Ruggeri A, Grisan E, Jaroszewski J (2005) A new system for the automatic estimation of endothelial cell density in donor corneas. Br J Ophthalmol 89(3):306–311
Hiroyasu T, Sekiya S, Nunokawa S, et al. (2013), Extracting rules for cell segmentation in corneal endothelial cell images using GP. In: IEEE International Conference on Systems, Man, and Cybernetics, pp. 1811–1816
Foracchia M, Ruggeri A, (2000), Cell contour detection in corneal endothelium in-vivo microscopy. In: Proceedings of the 22nd Annual International Conference of the IEEE Engineering in Medicine and Biology Society (Cat. No. 00CH37143), vol. 2, pp. 1033–1035
Vigueras-Guillen JP, van Rooij J, Engel A et al (2020) Deep learning for assessing the corneal endothelium from specular microscopy images up to 1 year after ultrathin-DSAEK surgery. Translat Vision Sci Technol 9(2):49–49
Daniel MC, Atzrodt L, Bucher F et al (2019) Automated segmentation of the corneal endothelium in a large set of ‘real-world’specular microscopy images using the u-net architecture. Sci Rep 9(1):1–7
Kolluru C, Benetz BA, Joseph N, et al. (2019), Machine learning for segmenting cells in corneal endothelium images. In: Medical Imaging 2019: Computer-Aided Diagnosis, vol. 10950, p. 109504
Nurzynska K (2018) Deep learning as a tool for automatic segmentation of corneal endothelium images. Symmetry 10(3):60
Vigueras-Guillen JP, Sari B, Goes SF et al (2019) Fully convolutional architecture vs sliding-window CNN for corneal endothelium cell segmentation. BMC Biomed Eng 1(1):1–16
Ronneberger O, Fischer P, Brox T (2015) U-net: Convolutional networks for biomedical image segmentation. International Conference on Medical image computing and computer-assisted intervention (pp. 234-241). Springer, Cham
Badrinarayanan V, Kendall A, Cipolla R (2017) Segnet: a deep convolutional encoder-decoder architecture for image segmentation. IEEE Trans Pattern Anal Mach Intell 39(12):2481–2495
Imre L, Nagymihaly A (2001) Reliability and reproducibility of corneal endothelial image analysis by in vivo confocal microscopy. Graefes Arch Clin Exp Ophthalmol 239(5):356–360
Piorkowski A, Nurzynska K, Boldak C et al (2015) Selected aspects of corneal endothelial segmentation quality. J Med Informat Technol:24
Sharif MS, Qahwaji R, Shahamatnia E, etal (2015) An efficient intelligent analysis system for confocal corneal endothelium images. Comput Methods Prog Biomed 122(3):421–436
Fabijanska A (2019) Automatic segmentation of corneal endothelial cells from microscopy images. Biomed Signal Proc Cont 47:145–158
Shannon CE (1948) A mathematical theory of communication. Bell Syst Tech J 27(3):379–423
Serra J (1983) Image Analysis and Mathematical Morphology. Academic Press Inc
Berzins V (1984) Accuracy of Laplacian edge detectors. Comput Vision Graphics Image Proc 27(2):195–210
Sobel I, Feldman G (1968) A 3x3 isotropic gradient operator for image processing. a talk at the. Stanford Artificial Project 271–272
Canny J (1986) A computational approach to edge detection. IEEE Trans Pattern Anal Mach Intell 6:679–698
Beucher S (1994 Watershed, hierarchical segmentation and waterfall algorithm. In: Mathematical Morphology and Its Applications to Image Processing, pp. 69–76. Springer
Funding
The research is supported by National Science Foundation.
Author information
Authors and Affiliations
Contributions
Conceptualization: Ranit Karmakar. Methodology: Ranit Karmakar. Formal analysis and investigation: Ranit Karmakar, Saeid Nooshabadi, Allen Eghrari. Writing—original draft preparation: Ranit Karmakar. Writing—review and editing: Saeid Nooshabadi, Allen Eghrari. Funding acquisition: Saeid Nooshabadi. Resources: Allen Eghrari. Supervision: Saeid Nooshabadi, Allen Eghrari.
Corresponding author
Ethics declarations
Ethics approval
The data used in this research was collected under IRB number IRB00039261 at Johns Hopkins University. All the data was deidentified before sharing with the researchers at Michigan Technological University; hence, they received a waiver for the analysis work.
Consent to participate
Patients were asked for their consent for research.
Consent for publication
Patients were asked for their consent for publication.
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
The original version of this article was revised. The IRB number in the Ethics approval section is now corrected.
Supplementary Information
Supplemental Figure 1.
Bland-Altman plot of difference in measured ECD between the manual flex-center and proposed algorithm measurements demonstrates minimal bias of 8 cells/mm2, approximately 0.3% error. (PNG 959 kb)
Supplemental Figure 2.
Distribution plot of the manual analysis, proposed algorithm and Auto Tracer for CD, HEX, and CV that shows the performance difference between Auto Tracer and proposed algorithm compared to the manual analysis. (PNG 768 kb)
Supplemental Figure 3.
Example images a original image, b region of interest demarcation, c feature-enhanced image, d cell segmentation, e segmented image with selected cells marked in red. (PNG 17574 kb)
Rights and permissions
About this article
Cite this article
Karmakar, R., Nooshabadi, S. & Eghrari, A. An automatic approach for cell detection and segmentation of corneal endothelium in specular microscope. Graefes Arch Clin Exp Ophthalmol 260, 1215–1224 (2022). https://doi.org/10.1007/s00417-021-05483-8
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00417-021-05483-8