Structure-Function Correlation of Deep-Learning Quantified Ellipsoid Zone and Retinal Pigment Epithelium Loss and Microperimetry in Geographic Atrophy

AI Generated Summary*

Microperimetric pointwise sensitivity was related to deep learning quantified OCT biomarkers in 20 eyes of 20 prospectively examined geographic atrophy patients (mean age 77). Two runs each on photopic MP-3 and mesopic MAIA with a 45-point grid gave 3600 stimuli co-registered to 1940 Spectralis B-scans. In multivariable mixed modelling at 0 degrees eccentricity, sensitivity was 2.81 dB lower with ellipsoid zone loss alone and 10.03 dB lower where RPE loss also occurred, versus non-atrophic areas; deficits grew with eccentricity yet remained significant throughout. Univariately, sensitivity increased 0.34 dB per micrometre of ellipsoid zone thickness, and drusen and hyperreflective foci volumes, but not age, were linked to lower values. Only 5% of points were absolute scotomas; analyses were deemed exploratory.

*This summary was generated by AI and is published unedited. Oku does not alter these summaries. It may contain errors or omissions and is provided for general informational purposes only. Oku does not guarantee its accuracy, completeness, or reliability. For authoritative information, please refer to the original, peer-reviewed article.