Artificial Intelligence Mapping of Structure to Function in Glaucoma

Abstract

Purpose:To develop an artificial intelligence (AI)-based structure-function (SF) map relating retinal nerve fiber layer (RNFL) damage on spectral domain optical coherence tomography (SDOCT) to functional loss on standard automated perimetry (SAP). Methods:The study included 26,499 pairs of SAP and SDOCT from 15,173 eyes of 8878 patients with glaucoma or suspected of having the disease extracted from the Duke Glaucoma Registry. The data set was randomly divided at the patient level in training and test sets. A convolutional neural network (CNN) was initially trained and validated to predict the 52 sensitivity threshold points of the 24-2 SAP from the 768 RNFL thickness points of the SDOCT peripapillary scan. Simulated localized RNFL defects of varied locations and depths were created by modifying the normal average peripapillary RNFL profile. The simulated profiles were then fed to the previously trained CNN, and the topographic SF relationships between structural defects and SAP functional losses were investigated. Results:The CNN predictions had an average correlation coefficient of 0.60 (P < 0.001) with the measured values from SAP and a mean absolute error of 4.25 dB. Simulated RNFL defects led to well-defined arcuate or paracentral visual field losses in the opposite hemifield, which varied according to the location and depth of the simulations. Conclusions:A CNN was capable of predicting SAP sensitivity thresholds from SDOCT RNFL thickness measurements and generate an SF map from simulated defects.Translational Relevance: AI-based SF map improves the understanding of how SDOCT losses translate into detectable SAP damage.

RetiNerveNet: using recursive deep learning to estimate pointwise 24-2 visual field data based on retinal structure

“…We propose RetiNerveNet, a deep fully convolutional neural architecture for obtaining estimates of SAP visual field values based on RNFL thickness values obtained from the more objective SDOCT tests. Unlike existing works with similar aim 9 , 10 , 13 – 15 , 17 , 18 , we postulate that building our network to mimic the arcuate structure of the axons of the retinal ganglion cells can help improve performance for this task. The fact that the proposed architecture performs better than a number of baselines in Table 2 seems to corroborate our hypothesis.…”

“…Some existing studies 13 , 14 , have used such deep learning techniques to estimate SAP summary metrics like Mean Deviation (MD, a weighted average of the age-corrected visual field values) using information acquired with SDOCT. Other works 15 – 17 , more closely related to our own, have attempted to estimate pointwise sensitivities for all the locations tested by SAP, based on the SDOCT thickness values. Some studies 18 , 19 have even attempted to predict the SAP sensitivities based on the raw images obtained from SDOCT.…”

An Artificial Intelligence Approach to Assess Spatial Patterns of Retinal Nerve Fiber Layer Thickness Maps in Glaucoma

“…Several studies have used deep learning techniques (i.e. deep neural network) 14 , 16 , 17 and a classical unsupervised AI method termed PCA 15 to quantitatively assess the high dimensional RNFLT map data. Compared with the “black box” deep learning approaches, the unsupervised AI methods typically aim to determine representative patterns from the data with reduced dimensionality to facilitate clinical assessment.…”

“…[12][13][14][15] To quantitatively assess the high dimensional RNFLT map data (e.g. 225 by 225 pixels for Cirrus OCT), various artificial intelligence (AI) techniques [14][15][16][17] have been developed. 12 For example, principal component analysis (PCA), a classical unsupervised AI method for data dimensionality reduction, 18,19 has been applied to determine 10 RNFLT patterns from the RNFLT maps.…”

Diagnostik von Erkrankungen des Sehnervenkopfes in Zeiten von künstlicher Intelligenz und Big Data

“…konnten eine KI-basierte Zuordnung von strukturellen OCT-RNFL-Schäden zu Gesichtsfelddefekten bei einem Glaukom entwickeln. Dies ermöglicht ein tieferes Verständnis des Zusammenhanges von Struktur und Funktion und kann im klinischen Alltag bei der Beurteilung von RNFL-Defekten helfen [ 34 ].…”