A fully automated cell segmentation and morphometric parameter system for quantifying corneal endothelial cell morphology

Al-Fahdawi, Shumoos; Qahwaji, Rami; Al-Waisy, Alaa S.; Ipson, Stanley S.; Ferdousi, Maryam; Malik, Rayaz A.; Brahma, Arun

doi:10.1016/j.cmpb.2018.03.015

Cited by 37 publications

(32 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We have previously demonstrated a reduction in corneal endothelial cell density in patients with diabetes [42, 43]. We have also recently developed an automated image analysis system to quantify corneal endothelial cell morphology and shown reduced corneal endothelial cell density and hypertrophy in patients with diabetes [24]. Given that we found corneal nerve and endothelial cell abnormalities in patients with TIA and minor stroke, we assessed for associations with cerebrovascular reactivity and risk factors for stroke.…”

Section: Discussionmentioning

confidence: 99%

“…Corneal nerve fiber density (CNFD), corneal nerve branch density (CNBD), corneal nerve fiber length (CNFL) and corneal nerve fiber tortuosity (CNFT) were analysed manually using CCMetrics (M. A. Dabbah, ISBE, University of Manchester, Manchester, United Kingdom) [12] and the investigator was blinded to the diagnosis. Corneal endothelial cell density, area, perimeter and degree of polymegathism (cell size variability) and pleomorphism (cell shape variability) were quantified using automated CEAS software [24].…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Corneal nerve and endothelial cell damage in patients with transient ischemic attack and minor ischemic stroke

et al. 2019

Self Cite

View full text Add to dashboard Cite

ObjectiveTo determine if corneal confocal microscopy can identify corneal nerve and endothelial cell abnormalities and may be useful in the prognostication of patients with transient ischemic attack [1] or minor ischemic stroke (IS).MethodsThirty-six patients admitted with TIA (n = 14) or minor IS (n = 22) underwent transcranial Doppler evaluation and corneal confocal microscopy and were compared with 18 healthy controls.ResultsCorneal nerve fiber density (P = 0.002), branch density (P = 0.004) and fiber length (P = 0.004) were significantly lower in patients with TIA or minor IS compared to controls, with no difference between patients with TIA and minor IS. Endothelial cell density (P = 0.003) was lower and endothelial cell area (P = 0.003) and perimeter (P = 0.006) were significantly higher in patients with TIA or minor IS compared to controls, with no difference between patients with TIA and minor IS. There were no differences in corneal nerve or endothelial cell morphology between patients with and without abnormal cerebrovascular reactivity. HbA1c was independently associated with CNFL, and endothelial cell polymegathism and pleomorphism were associated with both HbA1c and total cholesterol.ConclusionCorneal confocal microscopy identifies corneal nerve fiber loss and endothelial cell abnormalities in patients with TIA and minor IS and independent associations with HbA1c and cholesterol.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Corneal nerve and endothelial cell damage in patients with transient ischemic attack and minor ischemic stroke

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…The image analysis was performed blindly without the investigator being aware of whether the images were from a control subject or patient with stroke. Each image was exported to a real-time automated image analysis system (Corneal Endothelium Analysis System (CEAS)) 50 . A central region of interest (ROI) was traced for each image to identify the optimal area for quantification, avoiding peripheral darker areas.…”

Section: Methodsmentioning

confidence: 99%

Corneal Confocal Microscopy detects a Reduction in Corneal Endothelial Cells and Nerve Fibres in Patients with Acute Ischemic Stroke

Khan

Kamran

Akhtar

et al. 2018

Sci Rep

Self Cite

View full text Add to dashboard Cite

Endothelial dysfunction and damage underlie cerebrovascular disease and ischemic stroke. We undertook corneal confocal microscopy (CCM) to quantify corneal endothelial cell and nerve morphology in 146 patients with an acute ischemic stroke and 18 age-matched healthy control participants. Corneal endothelial cell density was lower (P < 0.001) and endothelial cell area (P < 0.001) and perimeter (P < 0.001) were higher, whilst corneal nerve fibre density (P < 0.001), corneal nerve branch density (P < 0.001) and corneal nerve fibre length (P = 0.001) were lower in patients with acute ischemic stroke compared to controls. Corneal endothelial cell density, cell area and cell perimeter correlated with corneal nerve fiber density (P = 0.033, P = 0.014, P = 0.011) and length (P = 0.017, P = 0.013, P = 0.008), respectively. Multiple linear regression analysis showed a significant independent association between corneal endothelial cell density, area and perimeter with acute ischemic stroke and triglycerides. CCM is a rapid non-invasive ophthalmic imaging technique, which could be used to identify patients at risk of acute ischemic stroke.

show abstract

“…Moreover, many of them suggested the necessity of user interaction to correct errors. In contrast, new clinically applicable methods have been proposed in recent years: Foracchia and Ruggeri [ 21 ] developed an algorithm based on Bayesian shape models, which later evolved into a genetic algorithm by Scarpa and Ruggeri [ 22 ]; Sharif et al [ 23 ] developed a hybrid model based on a combination of an active contour model (snakes) and a particle swarm optimization approach; Habrat et al [ 24 ] proposed an algorithm based on directional filters, which was clinically evaluated along with other methods [ 25 ]; Al-Fahdawi et al [ 26 ] suggested a method based on the watershed algorithm and Voronoi tessellations; Selig et al [ 27 ] employed Fourier analysis and the seeded watershed algorithm in a stochastic manner to segment confocal images; and Vigueras-Guillén et al [ 28 ] proposed a classifier-driven method to generate an accurate segmentation from an oversegmented image, using Selig et al’s approach [ 27 ] to generate the oversegmentation. Among these methods, the ones including a comparison with their respective microscope’s estimates were significantly more accurate, yet some mistakes were still present.…”

Section: Introductionmentioning

confidence: 99%

Fully convolutional architecture vs sliding-window CNN for corneal endothelium cell segmentation

et al. 2019

View full text Add to dashboard Cite

Background: Corneal endothelium (CE) images provide valuable clinical information regarding the health state of the cornea. Computation of the clinical morphometric parameters requires the segmentation of endothelial cell images. Current techniques to image the endothelium in vivo deliver low quality images, which makes automatic segmentation a complicated task. Here, we present two convolutional neural networks (CNN) to segment CE images: a global fully convolutional approach based on U-net, and a local sliding-window network (SW-net). We propose to use probabilistic labels instead of binary, we evaluate a preprocessing method to enhance the contrast of images, and we introduce a postprocessing method based on Fourier analysis and watershed to convert the CNN output images into the final cell segmentation. Both methods are applied to 50 images acquired with an SP-1P Topcon specular microscope. Estimates are compared against a manual delineation made by a trained observer. Results: U-net (AUC = 0.9938) yields slightly sharper, clearer images than SW-net (AUC = 0.9921). After postprocessing, U-net obtains a DICE = 0.981 and a MHD = 0.22 (modified Hausdorff distance), whereas SW-net yields a DICE = 0.978 and a MHD = 0.30. U-net generates a wrong cell segmentation in only 0.48% of the cells, versus 0.92% for the SW-net. U-net achieves statistically significant better precision and accuracy than both, Topcon and SW-net, for the estimates of three clinical parameters: cell density (ECD), polymegethism (CV), and pleomorphism (HEX). The mean relative error in U-net for the parameters is 0.4% in ECD, 2.8% in CV, and 1.3% in HEX. The computation time to segment an image and estimate the parameters is barely a few seconds. Conclusions: Both methods presented here provide a statistically significant improvement over the state of the art. U-net has reached the smallest error rate. We suggest a segmentation refinement based on our previous work to further improve the performance.

show abstract

A fully automated cell segmentation and morphometric parameter system for quantifying corneal endothelial cell morphology

Abstract: AS et al (2018) A fully automated cell segmentation and morphometric parameter system for quantifying corneal endothelial cell morphology. Computer Methods and Programs in Biomedicine. 160: 11-23. Rights

Cited by 37 publications

References 36 publications

Corneal nerve and endothelial cell damage in patients with transient ischemic attack and minor ischemic stroke

Corneal nerve and endothelial cell damage in patients with transient ischemic attack and minor ischemic stroke

Corneal Confocal Microscopy detects a Reduction in Corneal Endothelial Cells and Nerve Fibres in Patients with Acute Ischemic Stroke

Fully convolutional architecture vs sliding-window CNN for corneal endothelium cell segmentation

Contact Info

Product

Resources

About