Detection of Progressive Glaucomatous Optic Nerve Damage on Fundus Photographs with Deep Learning.

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. Retrospective cohort study. 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. 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. The relationship between change in predicted RNFL thickness from photographs and change in SD OCT RNFL thickness over time. 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%). 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.

Medeiros FA ，Jammal AA ，Mariottoni EB 《-》

From Machine to Machine: An OCT-Trained Deep Learning Algorithm for Objective Quantification of Glaucomatous Damage in Fundus Photographs.

Previous approaches using deep learning (DL) algorithms to classify glaucomatous damage on fundus photographs have been limited by the requirement for human labeling of a reference training set. We propose a new approach using quantitative spectral-domain (SD) OCT data to train a DL algorithm to quantify glaucomatous structural damage on optic disc photographs. Cross-sectional study. A total of 32 820 pairs of optic disc photographs and SD OCT retinal nerve fiber layer (RNFL) scans from 2312 eyes of 1198 participants. The sample was divided randomly into validation plus training (80%) and test (20%) sets, with randomization performed at the patient level. A DL convolutional neural network was trained to assess optic disc photographs and predict SD OCT average RNFL thickness. The DL algorithm performance was evaluated in the test sample by evaluating correlation and agreement between the predictions and actual SD OCT measurements. We also assessed the ability to discriminate eyes with glaucomatous visual field loss from healthy eyes with area under the receiver operating characteristic (ROC) curves. The mean prediction of average RNFL thickness from all 6292 optic disc photographs in the test set was 83.3±14.5 μm, whereas the mean average RNFL thickness from all corresponding SD OCT scans was 82.5±16.8 μm (P = 0.164). There was a very strong correlation between predicted and observed RNFL thickness values (Pearson r = 0.832; R2 = 69.3%; P < 0.001), with mean absolute error of the predictions of 7.39 μm. The area under the ROC curves for discriminating glaucomatous from healthy eyes with the DL predictions and actual SD OCT average RNFL thickness measurements were 0.944 (95% confidence interval [CI], 0.912-0.966) and 0.940 (95% CI, 0.902-0.966), respectively (P = 0.724). We introduced a novel DL approach to assess fundus photographs and provide quantitative information about the amount of neural damage that can be used to diagnose and stage glaucoma. In addition, training neural networks to predict SD OCT data objectively represents a new approach that overcomes limitations of human labeling and could be useful in other areas of ophthalmology.

Medeiros FA ，Jammal AA ，Thompson AC 《-》

Three-dimensional imaging of the macular retinal nerve fiber layer in glaucoma with spectral-domain optical coherence tomography.

Sakamoto A ，Hangai M ，Nukada M ，Nakanishi H ，Mori S ，Kotera Y ，Inoue R ，Yoshimura N ... -  《-》

A Deep Learning Algorithm to Quantify Neuroretinal Rim Loss From Optic Disc Photographs.

To train a deep learning (DL) algorithm that quantifies glaucomatous neuroretinal damage on fundus photographs using the minimum rim width relative to Bruch membrane opening (BMO-MRW) from spectral-domain optical coherence tomography (SDOCT). Cross-sectional study. A total of 9282 pairs of optic disc photographs and SDOCT optic nerve head scans from 927 eyes of 490 subjects were randomly divided into the validation plus training (80%) and test sets (20%). A DL convolutional neural network was trained to predict the SDOCT BMO-MRW global and sector values when evaluating optic disc photographs. The predictions of the DL network were compared to the actual SDOCT measurements. The area under the receiver operating curve (AUC) was used to evaluate the ability of the network to discriminate glaucomatous visual field loss from normal eyes. The DL predictions of global BMO-MRW from all optic disc photographs in the test set (mean ± standard deviation [SD]: 228.8 ± 63.1 μm) were highly correlated with the observed values from SDOCT (mean ± SD: 226.0 ± 73.8 μm) (Pearson's r = 0.88; R2 = 77%; P < .001), with mean absolute error of the predictions of 27.8 μm. The AUCs for discriminating glaucomatous from healthy eyes with the DL predictions and actual SDOCT global BMO-MRW measurements were 0.945 (95% confidence interval [CI]: 0.874-0.980) and 0.933 (95% CI: 0.856-0.975), respectively (P = .587). A DL network can be trained to quantify the amount of neuroretinal damage on optic disc photographs using SDOCT BMO-MRW as a reference. This algorithm showed high accuracy for glaucoma detection, and may potentially eliminate the need for human gradings of disc photographs.

Thompson AC ，Jammal AA ，Medeiros FA 《-》

Ability of cirrus high-definition spectral-domain optical coherence tomography clock-hour, deviation, and thickness maps in detecting photographic retinal nerve fiber layer abnormalities.

To investigate the ability of clock-hour, deviation, and thickness maps of Cirrus high-definition spectral-domain optical coherence tomography (HD-OCT) in detecting retinal nerve fiber layer (RNFL) defects identified in red-free fundus photographs in eyes with early glaucoma (mean deviation >-6.0 dB). Cross-sectional study. Two hundred ninety-five eyes with glaucomatous RNFL defects with clear margins observed in red-free fundus photographs and 200 age-, sex-, and refractive error-matched healthy eyes were enrolled. The width and location of RNFL defects were evaluated by using the red-free fundus photograph. When a RNFL defect detected by red-free fundus photograph did not present as (1) yellow/red codes in the clock-hour map, (2) yellow/red pixels in the deviation map, or (3) blue/black areas in the thickness map, the event was classified as a misidentification of a photographic RNFL defect by Cirrus HD-OCT. In healthy eyes, the presence of false-positive RNFL color codes of Cirrus HD-OCT maps was investigated. The prevalence of and factors associated with the (1) misidentification of photographic RNFL defects by Cirrus HD-OCT in eyes with glaucoma and (2) false-positive RNFL color codes of Cirrus HD-OCT maps in healthy eyes were assessed. Among the 295 red-free fundus photographic RNFL defects from 295 eyes with glaucoma, 83 (28.1%), 27 (9.2%), and 0 (0%) defects were misidentified in the clock-hour, deviation, and thickness maps of Cirrus HD-OCT, respectively. Fifty-six defects (19.0%) were misidentified only in the clock-hour map and 27 (9.2%) in both the clock-hour and deviation maps. The misidentification of photographic RNFL defects by Cirrus HD-OCT was associated with a narrower width and a temporal location of RNFL defects (P<0.05). Among the 200 healthy eyes, 25 (12.5%), 30 (15.0%), and 12 (6.0%) eyes had false-positive RNFL color codes in clock-hour, deviation, and thickness maps of Cirrus HD-OCT, respectively. Among the clock-hour, deviation, and thickness maps obtained with Cirrus HD-OCT, the thickness map showed the best diagnostic ability in detecting photographic RNFL defects. The RNFL thickness map may be a useful tool for the detection of RNFL defects in eyes with early glaucoma.

Hwang YH ，Kim YY ，Kim HK ，Sohn YH ... -  《-》