Glaucoma Classification Through SSVEP-Derived ON- and OFF-Pathway Features.

Area of Science:

• Ophthalmology
• Neuroscience
• Biomedical Engineering

Background:

• Glaucoma is a leading cause of irreversible blindness worldwide.
• Early detection and diagnosis are critical for managing glaucoma and preventing vision loss.
• Steady-state visually evoked potentials (SSVEPs) are electrophysiological measures of visual pathway function.

Purpose of the Study:

• To evaluate the relative contribution of amplitude and phase of ON- and OFF-pathway SSVEPs in classifying glaucoma patients.
• To compare the diagnostic utility of different SSVEP features for glaucoma detection.

Main Methods:

• SSVEPs were recorded using sawtooth luminance increments (ON-biasing) and decrements (OFF-biasing) at 2.73 Hz.
• Logistic regression models were trained on SSVEP data from 98 glaucoma patients and 71 controls.
• Classification performance was assessed using amplitude and phase features from ON- and OFF-pathway stimulation.

Main Results:

• Overall classification accuracy was moderate (AUC ≈ 0.7).
• Signal phase features significantly outperformed amplitude features for glaucoma classification.
• OFF-pathway biasing showed a statistically significant improvement with amplitude features, but this was not observed in low signal-to-noise data.

Conclusions:

• Signal phase of SSVEPs provides valuable information for glaucoma classification, often overlooked in clinical applications.
• OFF-pathway amplitude features may be less affected by low signal-to-noise ratios.
• Electrophysiological measures of visual signal delay are important for developing future glaucoma diagnostic tools.