Deep Learning Analysis of Widefield Cornea Endothelial Imaging in Fuchs Dystrophy

IF 4.6 Q1 OPHTHALMOLOGY

Ophthalmology science Pub Date : 2025-08-14 DOI:10.1016/j.xops.2025.100914

Kai Yuan Tey MBBS , Brian Juin Hsein Lee MBBS , Clarissa Ng MBBS , Qiu Ying Wong , Satish K. Panda PhD , Amrit Dash , Jipson Wong , Ezekiel Ze Ken Cheong MD , Jodhbir S. Mehta FRCS(Ed), PhD , Leopold Schmeterer MSc, PhD , Khin Yadanar Win PhD , Damon Wong PhD , Marcus Ang FRCS(Ed), PhD

{"title":"Deep Learning Analysis of Widefield Cornea Endothelial Imaging in Fuchs Dystrophy","authors":"Kai Yuan Tey MBBS , Brian Juin Hsein Lee MBBS , Clarissa Ng MBBS , Qiu Ying Wong , Satish K. Panda PhD , Amrit Dash , Jipson Wong , Ezekiel Ze Ken Cheong MD , Jodhbir S. Mehta FRCS(Ed), PhD , Leopold Schmeterer MSc, PhD , Khin Yadanar Win PhD , Damon Wong PhD , Marcus Ang FRCS(Ed), PhD","doi":"10.1016/j.xops.2025.100914","DOIUrl":null,"url":null,"abstract":"<div><h3>Purpose</h3><div>To evaluate the use of a deep learning network (DLN) in analyzing widefield specular microscopy (WFSM) images in eyes with Fuchs endothelial corneal dystrophy (FECD).</div></div><div><h3>Design</h3><div>Cross-sectional clinical observational study.</div></div><div><h3>Participants</h3><div>A total of 1839 images were obtained via WFSM imaging (CEM-530, Nidek Co Ltd) performed on 155 FECD eyes. A separate data set comprising images from 50 FECD eyes and 50 control eyes (70% training, 30% validation) was used for DLN training, which was tested on 354 images from 55 eyes from varying regions (central, paracentral, and peripheral).</div></div><div><h3>Methods</h3><div>Images were graded based on a standardized quality score. Central images were compared with paracentral and peripheral images in terms of quality and morphometric parameters: endothelial cell density (ECD), coefficient of variation (CV), and hexagonality (HEX). A U-Net-based DLN was developed and trained using the separate data set and then tested on an external longitudinal data set (baseline and 1 month). Segmentation accuracy between DLN and manual analysis was compared using the Sørensen–Dice coefficient. Morphometric outcomes (ECD, HEX, and CV) were analyzed using paired <em>t</em> tests.</div></div><div><h3>Main Outcome Measures</h3><div>Intergrader agreement for image quality and FECD severity; comparison of DLN-derived ECD with manual analysis.</div></div><div><h3>Results</h3><div>Strong intergrader agreement was observed for both image quality (κ = 0.967, 95% confidence interval [CI]: 0.959–0.976) and FECD severity (κ = 0.891, 95% CI: 0.786–0.995). Endothelial cell density differences between paracentral/peripheral regions were significant in eyes without or with subclinical edema (<em>P</em> = 0.001–0.011). Deep learning network-based segmentation closely matched manual results (Dice coefficient = 0.86 ± 0.04). Central ECD values obtained via DLN were significantly higher than manual analysis (DLN: 2633.12 ± 1167.3 cells/mm<sup>2</sup> vs. manual: 1728.58 ± 805.69 cells/mm<sup>2</sup>, <em>P</em> < 0.001).</div></div><div><h3>Conclusions</h3><div>This study presents a novel application of deep learning for analyzing widefield corneal endothelial images. The integration of a progression visualization tool enhances interpretability, allowing efficient autoanalysis and organization of large WFSM data sets—streamlining workflows and addressing limitations of manual interpretation.</div></div><div><h3>Financial Disclosure(s)</h3><div>The authors have no proprietary or commercial interest in any materials discussed in this article.</div></div>","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"6 1","pages":"Article 100914"},"PeriodicalIF":4.6000,"publicationDate":"2025-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Ophthalmology science","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S266691452500212X","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"OPHTHALMOLOGY","Score":null,"Total":0}

引用次数: 0

Abstract

Purpose

To evaluate the use of a deep learning network (DLN) in analyzing widefield specular microscopy (WFSM) images in eyes with Fuchs endothelial corneal dystrophy (FECD).

Design

Cross-sectional clinical observational study.

Participants

A total of 1839 images were obtained via WFSM imaging (CEM-530, Nidek Co Ltd) performed on 155 FECD eyes. A separate data set comprising images from 50 FECD eyes and 50 control eyes (70% training, 30% validation) was used for DLN training, which was tested on 354 images from 55 eyes from varying regions (central, paracentral, and peripheral).

Methods

Images were graded based on a standardized quality score. Central images were compared with paracentral and peripheral images in terms of quality and morphometric parameters: endothelial cell density (ECD), coefficient of variation (CV), and hexagonality (HEX). A U-Net-based DLN was developed and trained using the separate data set and then tested on an external longitudinal data set (baseline and 1 month). Segmentation accuracy between DLN and manual analysis was compared using the Sørensen–Dice coefficient. Morphometric outcomes (ECD, HEX, and CV) were analyzed using paired t tests.

Main Outcome Measures

Intergrader agreement for image quality and FECD severity; comparison of DLN-derived ECD with manual analysis.

Results

Strong intergrader agreement was observed for both image quality (κ = 0.967, 95% confidence interval [CI]: 0.959–0.976) and FECD severity (κ = 0.891, 95% CI: 0.786–0.995). Endothelial cell density differences between paracentral/peripheral regions were significant in eyes without or with subclinical edema (P = 0.001–0.011). Deep learning network-based segmentation closely matched manual results (Dice coefficient = 0.86 ± 0.04). Central ECD values obtained via DLN were significantly higher than manual analysis (DLN: 2633.12 ± 1167.3 cells/mm² vs. manual: 1728.58 ± 805.69 cells/mm², P < 0.001).

Conclusions

This study presents a novel application of deep learning for analyzing widefield corneal endothelial images. The integration of a progression visualization tool enhances interpretability, allowing efficient autoanalysis and organization of large WFSM data sets—streamlining workflows and addressing limitations of manual interpretation.

Financial Disclosure(s)

The authors have no proprietary or commercial interest in any materials discussed in this article.

查看原文本刊更多论文

Fuchs营养不良大视场角膜内皮成像的深度学习分析

目的探讨深度学习网络（DLN）在分析Fuchs内皮性角膜营养不良（FECD）患者的广角镜面显微镜（WFSM）图像中的应用。设计横断面临床观察研究。参与者通过WFSM成像（CEM-530, Nidek Co Ltd）对155只FECD眼睛进行成像，共获得1839张图像。DLN训练使用了一个独立的数据集，包括来自50只feecd眼和50只对照眼（70%训练，30%验证）的图像，并对来自不同区域（中央、旁中心和外围）的55只眼的354张图像进行了测试。方法采用标准化质量评分法对图像进行分级。在质量和形态学参数方面，将中心图像与中心旁图像和周围图像进行比较：内皮细胞密度（ECD）、变异系数（CV）和六边形（HEX）。使用单独的数据集开发和训练基于u - net的DLN，然后在外部纵向数据集（基线和1个月）上进行测试。利用Sørensen-Dice系数比较DLN与人工分析的分割精度。形态计量学结果（ECD、HEX和CV）采用配对t检验进行分析。主要结果测量：图像质量和FECD严重程度的评分者一致；dln衍生ECD与手工分析的比较。结果图像质量（κ = 0.967, 95%可信区间[CI]: 0.959-0.976）和FECD严重程度（κ = 0.891, 95% CI: 0.786-0.995）的积分一致性较强。在无亚临床水肿或有亚临床水肿的眼睛中，中心旁/外周区域内皮细胞密度差异显著（P = 0.001-0.011）。基于深度学习网络的分割结果与人工分割结果非常接近（Dice系数= 0.86±0.04）。DLN获得的中心ECD值显著高于手工分析（DLN: 2633.12±1167.3 cells/mm2 vs.手工：1728.58±805.69 cells/mm2, P < 0.001）。本研究提出了一种新的深度学习应用于分析大视场角膜内皮图像。进度可视化工具的集成增强了可解释性，允许对大型WFSM数据集进行有效的自动分析和组织，从而简化了工作流程并解决了手动解释的局限性。作者在本文中讨论的任何材料中没有专有或商业利益。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊