Related Experiment Video
Updated: Jan 17, 2026
14:33
Morphometric Analyses of Retinal Sections
Published on: February 19, 2012
10.7K
Ophthalmic Segmentation and Analysis Software (OASIS): A Comprehensive Tool for Quantitative Evaluation of
Naomi Joseph1, Ved Shivade1, Jiawei Chen1
1Department of Biomedical Engineering, Case Western Reserve University, Cleveland, OH, USA.
Translational Vision Science & Technology
|September 15, 2025
Summary
Ophthalmic Segmentation and Analysis Software (OASIS) significantly speeds up meibomian gland dysfunction analysis from meibography images. This AI-powered tool reduces analysis time by 87%, offering objective and efficient evaluation for clinical practice and research.
Area of Science:
- Ophthalmology
- Medical Imaging
- Artificial Intelligence
Background:
- Meibomian gland dysfunction (MGD) analysis relies on time-consuming manual interpretation of meibography images.
- Current methods lack objective and efficient quantitative evaluation, hindering MGD progression insights.
Purpose of the Study:
- To develop Ophthalmic Segmentation and Analysis Software (OASIS) for automated meibography image analysis.
- To improve the efficiency and objectivity of MGD assessment using AI.
Main Methods:
- A natural history study collected 2,439 meibography images from 325 patients.
- OASIS utilized manual annotation and deep-learning models for eyelid, gland, and gland loss segmentation.
- The software incorporated image enhancement and clinical metric calculation tools.
Main Results:
- OASIS reduced MGD analysis time by 87% (under 3 minutes per patient).
- Accurate Pult meiboscale grading showed fair agreement between clinicians and software (kappa = 0.79).
Conclusions:
- OASIS streamlines meibography analysis, providing objective and efficient MGD evaluation.
- AI-driven tools like OASIS can reduce clinician workload while maintaining accuracy.
- OASIS facilitates the development of quantitative MGD biomarkers for clinical and research applications.
