Convex hull based neuro-retinal optic cup ellipse optimization in glaucoma diagnosis

Area of Science:

• Ophthalmology
• Medical Imaging
• Computational Biology

Background:

• Glaucoma is a leading cause of irreversible blindness worldwide.
• Accurate diagnosis relies on measuring the optic cup-to-disc ratio (CDR).
• Automating optic cup boundary detection is complex due to vessel interference.

Purpose of the Study:

• To develop and validate an improved algorithm for automatic optic cup boundary estimation.
• To enhance the accuracy of neuro-retinal optic cup-to-disc ratio (CDR) calculation.
• To assess the clinical performance of the new algorithm in large patient cohorts.

Main Methods:

• A novel Convex Hull based Neuro-Retinal Optic Cup Ellipse Optimization algorithm was developed.
• The algorithm's accuracy was initially tested on a dataset of 70 patients from the Singapore Eye Research Institute.
• Performance was benchmarked against the state-of-the-art ARGALI system.

Main Results:

• The new algorithm demonstrated a 43% improvement in root mean squared error compared to the ARGALI system.
• Initial validation showed high accuracy in estimating the optic cup boundary.
• A large-scale clinical evaluation involving 15,000 patients from Australia and Singapore was conducted.

Conclusions:

• The Convex Hull based algorithm offers a significant advancement in accurately calculating the optic cup boundary.
• This improved accuracy in CDR measurement can enhance glaucoma diagnosis.
• The algorithm shows strong potential for widespread clinical application in glaucoma screening and management.