RetiNerveNet: Using Recursive Deep Learning to Estimate Pointwise 24-2 Visual Field Data based on Retinal Structure

TL;DR

RetiNerveNet is a recursive deep learning model that estimates visual field data from retinal structure, potentially improving glaucoma diagnosis by leveraging OCT data to predict visual field loss more accurately.

Contribution

The paper introduces RetiNerveNet, a novel recursive neural network that estimates visual field data from retinal nerve fiber layer measurements, outperforming baseline methods.

Findings

More accurate than baselines in estimating visual field values.

Effective in predicting Mean Deviation scores.

Ensemble approach further enhances performance.

Abstract

Glaucoma is the leading cause of irreversible blindness in the world, affecting over 70 million people. The cumbersome Standard Automated Perimetry (SAP) test is most frequently used to detect visual loss due to glaucoma. Due to the SAP test's innate difficulty and its high test-retest variability, we propose the RetiNerveNet, a deep convolutional recursive neural network for obtaining estimates of the SAP visual field. RetiNerveNet uses information from the more objective Spectral-Domain Optical Coherence Tomography (SDOCT). RetiNerveNet attempts to trace-back the arcuate convergence of the retinal nerve fibers, starting from the Retinal Nerve Fiber Layer (RNFL) thickness around the optic disc, to estimate individual age-corrected 24-2 SAP values. Recursive passes through the proposed network sequentially yield estimates of the visual locations progressively farther from the optic disc. While all the methods used for our experiments exhibit lower performance for the advanced disease group, the proposed network is observed to be more accurate than all the baselines for estimating the individual visual field values. We further augment RetiNerveNet to additionally predict the SAP Mean Deviation values and also create an ensemble of RetiNerveNets that further improves the performance, by increasingly weighting-up underrepresented parts of the training data.