Related Experiment Video
Updated: Dec 11, 2025
Full-Field Optical Coherence Microscopy for Histology-Like Analysis of Stromal Features in Corneal Grafts
Published on: October 21, 2022
Measurement Reliability for Keratitis Morphology
Matthias F Kriegel1,2, Jessica Loo3, Sina Farsiu3,4
1Department of Ophthalmology and Visual Sciences, W.K. Kellogg Eye Center, University of Michigan, Ann Arbor, MI.
Purpose:
To evaluate the reliability of manual annotation when quantifying cornea anatomical and microbial keratitis (MK) morphological features on slit-lamp photography (SLP) images.
Methods:
Prospectively enrolled patients with MK underwent SLP at initial encounter at 2 academic eye hospitals. Patients who presented with an epithelial defect (ED) were eligible for analysis. Features, which included ED, corneal limbus (L), pupil (P), stromal infiltrate (SI), white blood cell (WBC) infiltration at the SI edge, and hypopyon (H), were annotated independently by 2 physicians on SLP images. Intraclass correlation coefficients (ICCs) were applied for reliability assessment; dice similarity coefficients (DSCs) were used to investigate the area overlap between readers.
Results:
Seventy-five MK patients with an ED received SLP. DSCs indicate good to fair annotation overlap between graders (L = 0.97, P = 0.80, ED = 0.94, SI = 0.82, H = 0.82, WBC = 0.83) and between repeat annotations by the same grader (L = 0.97, P = 0.81, ED = 0.94, SI = 0.85, H = 0.84, WBC = 0.82). ICC scores showed good intergrader (L = 0.98, P = 0.78, ED = 1.00, SI = 0.67, H = 0.97, WBC = 0.86) and intragrader (L = 0.99, P = 0.92, ED = 0.99, SI = 0.94, H = 0.99, WBC = 0.92) reliabilities. When reliability statistics were recalculated for annotated SI area in the subset of cases where both graders agreed WBC infiltration was present/absent, intergrader ICC improved to 0.91 and DSC improved to 0.86 and intragrader ICC remained the same, whereas DSC improved to 0.87.
Conclusions:
Manual annotation indicates usefulness of area quantification in the evaluation of MK. However, variability is intrinsic to the task. Thus, there is a need for optimization of annotation protocols. Future directions may include using multiple annotators per image or automated annotation software.
Insights
Manual annotation of microbial keratitis (MK) features on slit-lamp photography (SLP) shows good reliability for quantifying anatomical and morphological characteristics. Optimization of annotation protocols is recommended to minimize inherent variability in manual assessments.
Area of Science:
- Ophthalmology
- Medical Imaging
- Computational Biology
Background:
- Microbial keratitis (MK) is a serious eye infection requiring accurate assessment of anatomical and morphological features.
- Slit-lamp photography (SLP) is a key diagnostic tool, but manual annotation of its images can be subjective.
Purpose of the Study:
- To evaluate the reliability of manual annotation for quantifying corneal anatomical and MK morphological features on SLP images.
- To assess inter- and intra-grader variability in feature annotation.
Main Methods:
- Prospective enrollment of 75 MK patients with epithelial defects.
- Independent manual annotation of features (corneal limbus, pupil, epithelial defect, stromal infiltrate, white blood cell infiltration, hypopyon) on SLP images by two physicians.
- Reliability assessment using intraclass correlation coefficients (ICCs) and dice similarity coefficients (DSCs).
Main Results:
- Good to fair annotation overlap (DSCs ranging from 0.80 to 0.97) and high reliability (ICCs ranging from 0.67 to 1.00) were observed between graders and for repeat annotations.
- Recalculation for stromal infiltrate area, with agreement on white blood cell infiltration presence/absence, improved inter-grader ICC to 0.91 and DSC to 0.86.
Conclusions:
- Manual annotation is useful for quantifying MK features on SLP images.
- Inherent variability necessitates optimization of annotation protocols.
- Future directions include using multiple annotators or automated annotation software.
