Related Experiment Video
Updated: Jan 1, 2026
06:29
Comparing Objective Conjunctival Hyperemia Grading and the Ocular Surface Disease Index Score in Dry Eye Syndrome During COVID-19
Published on: May 25, 2022
2.7K
Validation of Computerized Quantification of Ocular Redness.
Ekaterina Sirazitdinova1, Marlies Gijs2, Christian J F Bertens2
1Uniklinik RWTH Aachen, Department of Medical Informatics, Aachen, Germany.
Translational Vision Science & Technology
|December 20, 2019
Summary
Computerized ocular redness quantification is feasible using automated methods. Automated segmentation and hue-saturation-value (HSV) color scoring offer consistent and objective results for clinical studies.
Area of Science:
- Ophthalmology
- Medical Imaging
- Computer Vision
Background:
- Ocular redness quantification is crucial for clinical studies.
- Manual assessment of ocular redness can be subjective and inconsistent.
- Objective and automated methods are needed for reliable quantification.
Purpose of the Study:
- To demonstrate the feasibility of computerized techniques for quantifying ocular redness in clinical settings.
- To develop an automatic and objective method for ocular redness assessment.
- To compare different segmentation and scoring methods for ocular redness.
Main Methods:
- Developed software for quantifying bulbar conjunctiva redness.
- Implemented manual and automatic sclera segmentation with automated alignment.
- Utilized redness scoring methods including CLAHE in RGB and HSV color spaces.
- Validated the pipeline based on segmentation reliability, precision, automation, and scoring method choice.
Main Results:
- Automated segmentation demonstrated the lowest variability, indicating robustness.
- Manual segmentation showed significant interobserver and intraobserver variability.
- Hue-saturation-value (HSV) based scoring methods were more consistent than RGB-based methods.
- RGB-based scoring was less sensitive in redness assessment.
Conclusions:
- Accurate ocular redness computation relies heavily on robust sclera segmentation.
- Automated segmentation provides consistent results, unlike subjective manual segmentation.
- HSV color space methods offer more consistent ocular redness scoring.
- Computerized quantification of ocular redness promises to objectify assessments in clinical care and trials.
