Focal edge association to glaucoma diagnosis

Area of Science:

• Ophthalmology
• Medical Imaging
• Computer Vision

Background:

• Glaucoma, an optic nerve disease, causes vision loss and requires accurate classification.
• Iridocorneal angle grading is crucial for glaucoma diagnosis, but manual methods are subjective and time-consuming.
• Distinguishing between open angle glaucoma and angle closure glaucoma is clinically significant.

Purpose of the Study:

• To develop and validate an automated method for iridocorneal angle grading using focal edge detection.
• To improve the objectivity and efficiency of glaucoma type classification.
• To establish a machine learning-based association between focal edges and angle grades.

Main Methods:

• Automated detection of the iris surface to identify focal regions and focal edges.
• Application of machine learning to correlate focal edges with iridocorneal angle grades.
• Utilizing a modified three-grade classification system for angle grading.

Main Results:

• The proposed automated method achieved high accuracy in classifying glaucoma types: 87.3% for open angle and 88.4% for closed angle.
• Accurate classification of angle closure subtypes was demonstrated, with 75.0% for grade 1 and 77.4% for grade 0.
• The focal edge detection method provides a quantitative and reproducible approach to angle grading.

Conclusions:

• Automated iridocorneal angle grading using focal edge detection is a viable alternative to manual grading.
• The proposed machine learning model enhances the accuracy and efficiency of glaucoma diagnosis.
• This technology has the potential to streamline clinical workflows and improve patient outcomes in glaucoma management.