Scholarlink

<p>The purpose of this study was to compare manual assessment of corneal nerve fiber length (CNFL) and dendritic cell (DC) density with an automated assessment method utilizing deep learning segmentation to perform rule-based density estimation. Corneal images were acquired using in vivo confocal microscopy (IVCM) from 100 participants with persistent ocular symptoms after mild COVID-19 (Group 1) and 30 controls without symptoms (Group 2). In total, 1,300 IVCM images were selected and manually annotated for CNFL, and 1,300 for DCs (with dendrites and without dendrites), using FIJI tools. The between-method difference in mean CNFL density was 0.2 <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(\hbox {mm/mm}^2\)</EquationSource> </InlineEquation> (95% CI: [0.09, 0.23]) for Group 1 and −0.2 <InlineEquation ID="IEq2"> <EquationSource Format="TEX">\(\hbox {mm/mm}^2\)</EquationSource> </InlineEquation> (95% CI: [−0.34, −0.10]) for Group 2. For Group 1, the mean difference for DCs with dendrites was −1.1 <InlineEquation ID="IEq3"> <EquationSource Format="TEX">\(\hbox {cells/mm}^2\)</EquationSource> </InlineEquation> (95% CI: [−1.78, −0.39]), and for DCs without dendrites it was −3.1 <InlineEquation ID="IEq4"> <EquationSource Format="TEX">\(\hbox {cells/mm}^2\)</EquationSource> </InlineEquation> (95% CI: [−5.1, −1.0]). For Group 2, the mean difference for DCs with dendrites was −1.0 <InlineEquation ID="IEq5"> <EquationSource Format="TEX">\(\hbox {cells/mm}^2\)</EquationSource> </InlineEquation> (95% CI: [−1.79, −0.27]), and for DCs without dendrites it was 0.3 <InlineEquation ID="IEq6"> <EquationSource Format="TEX">\(\hbox {cells/mm}^2\)</EquationSource> </InlineEquation> (95% CI: [−1.93, 2.60]). Both manual and automated methods showed significant between-group differences for CNFL (<i>p</i>=0.012 and <i>p</i>=0.034, respectively) and DC densities (<i>p</i>=0.005 and <i>p</i>=0.010). The automated approach performed comparably to manual assessment, supporting its potential for reliable, scalable analysis of CNFL and DC in IVCM images.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Deep learning-based segmentation and density estimation of corneal nerves and dendritic cells from In Vivo confocal microscopy images

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年1月13日

查看全文
  • Meichen Ji,
  • Yan Song,
  • Jenny Roth,
  • Ava Dashti,
  • Jorge Lazo,
  • Alisa Lincke,
  • António Filipe Teixeira Macedo,
  • Welf Löwe,
  • Neil Lagali

摘要

The purpose of this study was to compare manual assessment of corneal nerve fiber length (CNFL) and dendritic cell (DC) density with an automated assessment method utilizing deep learning segmentation to perform rule-based density estimation. Corneal images were acquired using in vivo confocal microscopy (IVCM) from 100 participants with persistent ocular symptoms after mild COVID-19 (Group 1) and 30 controls without symptoms (Group 2). In total, 1,300 IVCM images were selected and manually annotated for CNFL, and 1,300 for DCs (with dendrites and without dendrites), using FIJI tools. The between-method difference in mean CNFL density was 0.2 \(\hbox {mm/mm}^2\) (95% CI: [0.09, 0.23]) for Group 1 and −0.2 \(\hbox {mm/mm}^2\) (95% CI: [−0.34, −0.10]) for Group 2. For Group 1, the mean difference for DCs with dendrites was −1.1 \(\hbox {cells/mm}^2\) (95% CI: [−1.78, −0.39]), and for DCs without dendrites it was −3.1 \(\hbox {cells/mm}^2\) (95% CI: [−5.1, −1.0]). For Group 2, the mean difference for DCs with dendrites was −1.0 \(\hbox {cells/mm}^2\) (95% CI: [−1.79, −0.27]), and for DCs without dendrites it was 0.3 \(\hbox {cells/mm}^2\) (95% CI: [−1.93, 2.60]). Both manual and automated methods showed significant between-group differences for CNFL (p=0.012 and p=0.034, respectively) and DC densities (p=0.005 and p=0.010). The automated approach performed comparably to manual assessment, supporting its potential for reliable, scalable analysis of CNFL and DC in IVCM images.