Determinants of cone- and rod-function in geographic atrophy: AI-based structure-function correlation

Journal article

Purpose To investigate the association between retinal microstructure and cone- and rod-function in geographic atrophy (GA) secondary to age-related macular degeneration (AMD) using artificial-intelligence-(AI) algorithms.

Design Prospective, observational case series

Methods Forty-one eyes of 41 patients (75.8±8.4 years; 22 female) from a tertiary referral hospital were included (Directional-Spread-in-Geographic-Atrophy (DSGA) natural history study; NCT02051998). Mesopic, dark-adapted (DA) cyan and red sensitivity were assessed using fundus-controlled perimetry (“microperimetry”); retinal microstructure using spectral-domain optical-coherence-tomography (SD-OCT), fundus autofluorescence (FAF) and near-infrared-reflectance (IR) imaging. Layer-thicknesses and -intensities and FAF- and IR-intensities were extracted for each test-point. We evaluated the cross-validated mean absolute error (MAE) for random-forest-based predictions of retinal sensitivity with and without patient-specific training-data and the increase mean-squared error (%IncMSE) as measure of feature-importance.

Results Retinal sensitivity was predicted with a MAE of 4.64 dB for mesopic, 4.89 dB for DA cyan and 4.40 dB for DA red testing in absence of patient-specific data. Partial addition of patient-specific sensitivity data to the training sets decreased the MAE to 2.89 dB, 2.86 dB and 2.77 dB. For all three types of testing, the outer nuclear layer-thickness constituted the most important predictive feature (35.0, 42.22 and 53.74 %IncMSE). Spatially-resolved mapping of “inferred sensitivity” revealed regions with differential degrees of mesopic and DA cyan sensitivity loss outside of the GA lesions.

Conclusions “Inferred sensitivity” accurately reflected retinal function in patients with GA. Mapping of “inferred sensitivity” could facilitate monitoring of disease progression and serve as “quasi functional” surrogate outcome in clinical trials, especially in consideration of retinal regions beyond areas of GA.

Authors: Pfau, M; von der Emde, L; Dysli, C; Möller, PT; Thiele, S

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: Elsevier
• Journal: American Journal of Ophthalmology
• Volume: 217
• Pages: 162-173
• Publication date: 2020-04-11
• Acceptance date: 2020-04-03
• DOI: 10.1016/j.ajo.2020.04.003
• ISSN: 0002-9394
• Language: English
• Keywords: age-related macular degeneration; scotopic microperimetry; machine learning; microperimetry; FFR; geographic atrophy; random forest regression