Detecting glaucoma progression from localized rates of retinal changes in parametric and nonparametric statistical...

Area of Science:

• Ophthalmology
• Medical Imaging
• Glaucoma Research

Background:

• Glaucomatous progression detection is crucial for timely intervention.
• Current methods may have limitations in sensitivity or specificity.
• Pixelwise rates of retinal height changes (PixR) offer a novel approach.

Purpose of the Study:

• To evaluate three new PixR strategies for detecting glaucomatous progression.
• To compare the diagnostic accuracy of PixR methods against statistic image mapping (SIM).
• To reduce false-positive errors in glaucoma progression detection.

Main Methods:

• Compared PixR-NP cluster test (CT), PixR-NP single threshold test (STT), and PixR-P STT to SIM using Heidelberg Retina Tomograph.
• Included 36 progressing eyes, 210 non-progressing patient eyes, and 21 longitudinal normal eyes.
• Addressed multiple comparison issues using family-wise error rate (FWER) and k-FWER controls.

Main Results:

• PixR strategies demonstrated higher sensitivity in progressing eyes (75-81%) compared to SIM (39%).
• Specificity in normal eyes was comparable across methods (90%).
• PixR detected progression in some non-progressing eyes, correlating with early visual field changes.

Conclusions:

• PixR strategies offer improved sensitivity for detecting glaucomatous progression.
• PixR methods show potential to enhance diagnostic capabilities in glaucoma management.
• Longer follow-up is required to validate findings in non-progressing eyes.