Purpose To investigate whether predictions of retinal nerve fiber layer (RNFL) thickness obtained from a deep learning model applied to fundus photographs can detect progressive glaucomatous changes over time. Design Retrospective cohort study. Participants Eighty-six thousand one hundred twenty-three pairs of color fundus photographs and spectral-domain (SD) OCT images collected during 21 232 visits from 8831 eyes of 5529 patients with glaucoma or glaucoma suspects. Methods A deep learning convolutional neural network was trained to assess fundus photographs and to predict SD OCT global RNFL thickness measurements. The model then was tested on an independent sample of eyes that had longitudinal follow-up with both fundus photography and SD OCT. The ability to detect eyes that had statistically significant slopes of SD OCT change was assessed by receiver operating characteristic (ROC) curves. The repeatability of RNFL thickness predictions was investigated by measurements obtained from multiple photographs that had been acquired during the same day. Main Outcome Measures The relationship between change in predicted RNFL thickness from photographs and change in SD OCT RNFL thickness over time. Results The test sample consisted of 33 466 pairs of fundus photographs and SD OCT images collected during 7125 visits from 1147 eyes of 717 patients. Eyes in the test sample were followed up for an average of 5.3 ± 3.3 years, with an average of 6.2 ± 3.8 visits. A significant correlation was found between change over time in predicted and observed RNFL thickness ( r = 0.76; 95% confidence interval [CI], 0.70–0.80; P < 0.001). Retinal nerve fiber layer predictions showed an ROC curve area of 0.86 (95% CI, 0.83–0.88) to discriminate progressors from nonprogressors. For detecting fast progressors (slope faster than 2 μm/year), the ROC curve area was 0.96 (95% CI, 0.94–0.98), with a sensitivity of 97% for 80% specificity and 85% for 90% specificity. For photographs obtained at the same visit, the intraclass correlation coefficient was 0.946 (95% CI, 0.940–0.952), with a coefficient of variation of 3.2% (95% CI, 3.1%–3.3%). Conclusions A deep learning model was able to obtain objective and quantitative estimates of RNFL thickness that correlated well with SD OCT measurements and potentially could be used to monitor for glaucomatous changes over time.

IMPORTANCE Conventional segmentation of the retinal nerve fiber layer (RNFL) is prone to errors that may affect the accuracy of spectral-domain optical coherence tomography (SD-OCT) scans in detecting glaucomatous damage.OBJECTIVE To develop a segmentation-free deep learning (DL) algorithm for assessment of glaucomatous damage using the entire circle B-scan image from SD-OCT. DESIGN, SETTING, AND PARTICIPANTSThis cross-sectional study at a single institution used data from SD-OCT images of eyes with glaucoma (perimetric and preperimetric) and normal eyes. The data set was randomly split at the patient level into a training (50%), validation (20%), and test data set (30%). Data were collected from March 2008 to April 2019, and analysis began April 2018.EXPOSURES A convolutional neural network was trained to discriminate glaucomatous from normal eyes using the SD-OCT circle B-scan without segmentation lines. MAIN OUTCOMES AND MEASURESThe ability to discriminate glaucoma from healthy eyes was evaluated by comparing the area under the receiver operating characteristic curve and sensitivity at 80% or 95% specificity for the DL algorithm's predicted probability of glaucoma vs conventional RNFL thickness parameters given by SD-OCT software. The performance was also assessed in preperimetric glaucoma, as well as by visual field severity using Hodapp-Parrish-Anderson criteria.RESULTS A total of 20 806 SD-OCT images from 1154 eyes of 635 individuals (612 [53%] with glaucoma and 542 normal eyes [47%]) were included. The mean (SD) age at SD-OCT scan was 70.8 (10.4) years in individuals with glaucoma and 55.8 (14.1) years in controls. There were 187 women (53.3%) in the glaucoma group and 165 (59.8%) in the control group. Of 612 eyes with glaucoma, 432 (70.4%) had perimetric and 180 (29.6%) had preperimetric glaucoma. The DL algorithm had a significantly higher area under the receiver operating characteristic curve than global RNFL thickness (0.96 vs 0.87; difference = 0.08 [95% CI, 0.04-0.12]) and each RNFL thickness sector for discriminating between glaucoma and controls (all P < .001). At 95% specificity, the DL algorithm (81%; 95% CI, 64%-97%) was more sensitive than global RNFL thickness (67%; 95% CI, 58%-76%). The areas under the receiver operating characteristic curve were also significantly greater for the DL algorithm compared with RNFL thickness at each stage of disease, especially preperimetric and mild perimetric glaucoma. CONCLUSIONS AND RELEVANCEA segmentation-free DL algorithm performed better than conventional RNFL thickness parameters for diagnosing glaucomatous damage on OCT scans, especially in early disease. Future studies should investigate how such an approach contributes to diagnostic decisions when combined with other relevant clinical information, such as risk factors and perimetry results.

No abstract

No 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.