Predicting the Integrated Visual Field with Wide-Scan Optical Coherence Tomography in Glaucoma Patients

Area of Science:

• Ophthalmology
• Medical Imaging
• Glaucoma Research

Background:

• Standard visual field testing is crucial for glaucoma diagnosis and management.
• Optical coherence tomography (OCT) provides detailed structural information of the retina.
• Developing novel methods to predict visual field data from OCT is an active area of research.

Purpose of the Study:

• To calculate a predicted integrated visual field (IVF) using a novel method based on predicted monocular visual fields (MVFs).
• To derive MVFs from wide-scan OCT data.
• To validate the accuracy of the predicted IVF.

Main Methods:

• Visual field data from Humphrey Field Analyzer 24-2 program was used.
• OCT scans of the macula were analyzed for macular retinal nerve fiber layer (mRNFL), ganglion cell layer + inner plexiform layer (GCIPL), and GCC thickness.
• A support vector machine model was trained to predict MVFs from OCT measurements.
• The prediction model was validated using independent datasets for both MVF and IVF.

Main Results:

• The MVF prediction model showed varying prediction interval lengths for mRNFL, GCIPL, and GCC.
• Root mean squared errors between actual and predicted IVF sensitivity were within acceptable ranges across the 24 grid areas.
• Intraclass correlation coefficients indicated moderate agreement between actual and predicted IVF for mRNFL and GCC.

Conclusions:

• A predicted IVF can be accurately calculated from predicted MVFs derived using a new OCT-based method.
• This novel technique shows potential to enhance glaucoma management, particularly in patients who find standard visual field testing difficult.
• Further validation may establish this method as a valuable adjunct in clinical practice.