Spatial analyses of glaucomatous visual fields; a comparison with traditional visual field indices

Area of Science:

• Ophthalmology
• Medical Imaging
• Computational Vision

Background:

• Automated threshold visual field testing generates extensive data, posing interpretation challenges.
• Traditional indices like mean and standard deviation of threshold departures lack spatial context.
• Subjective visual field interpretation emphasizes spatial relationships, which is crucial for diagnosing conditions like glaucoma.

Purpose of the Study:

• To evaluate the efficacy of novel spatial analysis methods for automated visual field interpretation.
• To compare the diagnostic performance of spatial methods against traditional non-spatial indices in distinguishing normal from glaucomatous visual fields.
• To determine if spatial models enhance the accuracy of glaucoma detection.

Main Methods:

• Developed and applied two spatial analysis methods: regional up-down comparisons and arcuate cluster analysis.
• Utilized logistic regression analysis to discriminate between normal and glaucomatous visual fields using both spatial and non-spatial indices.
• Validated the spatial model's performance on an independent dataset, including eyes with and without large field defects.

Main Results:

• A spatial visual field model combining up-down differences and arcuate clusters achieved the best field classification.
• In an independent dataset, the spatial model demonstrated 87% sensitivity and 83% specificity.
• The best non-spatial model achieved 82% sensitivity and 80% specificity, indicating superior performance of the spatial approach.

Conclusions:

• Spatial visual field interpretation models significantly enhance the detection of glaucoma.
• Focusing on circumscribed field loss using spatial analysis offers superior diagnostic accuracy compared to non-spatial indices.
• These findings suggest a paradigm shift towards spatial analysis for more effective visual field interpretation in glaucoma management.