Signal-to-Noise Ratios for Structural and Functional Tests in Glaucoma

Area of Science:

• Ophthalmology
• Glaucoma research
• Medical imaging analysis

Background:

• Standard automated perimetry (SAP) for glaucoma diagnosis exhibits significant variability.
• Optical coherence tomography (OCT) offers potentially more repeatable structural measurements.
• Direct comparison of SAP and OCT for detecting glaucomatous changes is difficult due to differing units and dynamic ranges.

Purpose of the Study:

• To introduce a signal-to-noise ratio (SNR) analysis for comparing the repeatability of glaucoma detection methods.
• To compare the longitudinal SNR of OCT retinal nerve fiber layer thickness (RNFLT) and SAP mean deviation (MD).

Main Methods:

• Longitudinal data from 226 eyes of 130 glaucoma patients were analyzed.
• Retinal nerve fiber layer thickness (RNFLT) from OCT and mean deviation (MD) from SAP were linearly regressed against time.
• Signal was defined as the rate of change; noise was defined as the standard deviation of residuals. Individual and summary SNRs were calculated.

Main Results:

• OCT RNFLT demonstrated significantly better individual longitudinal signal-to-noise ratios compared to SAP MD (P < 0.0001).
• The summary SNR quantification was superior for OCT RNFLT (-1.35 y⁻¹) versus SAP MD (-0.74 y⁻¹).

Conclusions:

• Retinal nerve fiber layer thickness measured by OCT has a superior longitudinal signal-to-noise ratio over SAP mean deviation for glaucoma monitoring.
• The longitudinal SNR framework provides a robust method for comparing different diagnostic techniques, irrespective of scale or units.
• Reporting longitudinal SNR in glaucoma studies can enhance result interpretation and facilitate comparisons between methodologies.