Deep Learning Approaches Predict Glaucomatous Visual Field Damage from OCT Optic Nerve Head En Face Images and Retinal Nerve Fiber Layer Thickness Maps.

To develop and evaluate a deep learning system for differentiating between eyes with and without glaucomatous visual field damage (GVFD) and predicting the severity of GFVD from spectral domain OCT (SD OCT) optic nerve head images. Evaluation of a diagnostic technology. A total of 9765 visual field (VF) SD OCT pairs collected from 1194 participants with and without GVFD (1909 eyes). Deep learning models were trained to use SD OCT retinal nerve fiber layer (RNFL) thickness maps, RNFL en face images, and confocal scanning laser ophthalmoscopy (CSLO) images to identify eyes with GVFD and predict quantitative VF mean deviation (MD), pattern standard deviation (PSD), and mean VF sectoral pattern deviation (PD) from SD OCT data. Deep learning models were compared with mean RNFL thickness for identifying GVFD using area under the curve (AUC), sensitivity, and specificity. For predicting MD, PSD, and mean sectoral PD, models were evaluated using R2 and mean absolute error (MAE). In the independent test dataset, the deep learning models based on RNFL en face images achieved an AUC of 0.88 for identifying eyes with GVFD and 0.82 for detecting mild GVFD significantly (P < 0.001) better than using mean RNFL thickness measurements (AUC = 0.82 and 0.73, respectively). Deep learning models outperformed standard RNFL thickness measurements in predicting all quantitative VF metrics. In predicting MD, deep learning models based on RNFL en face images achieved an R2 of 0.70 and MAE of 2.5 decibels (dB) compared with 0.45 and 3.7 dB for RNFL thickness measurements. In predicting mean VF sectoral PD, deep learning models achieved high accuracy in the inferior nasal (R2 = 0.60) and superior nasal (R2 = 0.67) sectors, moderate accuracy in inferior (R2 = 0.26) and superior (R2 = 0.35) sectors, and lower accuracy in the central (R2 = 0.15) and temporal (R2 = 0.12) sectors. Deep learning models had high accuracy in identifying eyes with GFVD and predicting the severity of functional loss from SD OCT images. Accurately predicting the severity of GFVD from SD OCT imaging can help clinicians more effectively individualize the frequency of VF testing to the individual patient.

Christopher M ，Bowd C ，Belghith A ，Goldbaum MH ，Weinreb RN ，Fazio MA ，Girkin CA ，Liebmann JM ，Zangwill LM ... -  《-》

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 《-》

Deep Learning Estimation of 10-2 and 24-2 Visual Field Metrics Based on Thickness Maps from Macula OCT.

To develop deep learning (DL) systems estimating visual function from macula-centered spectral-domain (SD) OCT images. Evaluation of a diagnostic technology. A total of 2408 10-2 visual field (VF) SD OCT pairs and 2999 24-2 VF SD OCT pairs collected from 645 healthy and glaucoma subjects (1222 eyes). Deep learning models were trained on thickness maps from Spectralis macula SD OCT to estimate 10-2 and 24-2 VF mean deviation (MD) and pattern standard deviation (PSD). Individual and combined DL models were trained using thickness data from 6 layers (retinal nerve fiber layer [RNFL], ganglion cell layer [GCL], inner plexiform layer [IPL], ganglion cell-IPL [GCIPL], ganglion cell complex [GCC] and retina). Linear regression of mean layer thicknesses were used for comparison. Deep learning models were evaluated using R2 and mean absolute error (MAE) compared with 10-2 and 24-2 VF measurements. Combined DL models estimating 10-2 achieved R2 of 0.82 (95% confidence interval [CI], 0.68-0.89) for MD and 0.69 (95% CI, 0.55-0.81) for PSD and MAEs of 1.9 dB (95% CI, 1.6-2.4 dB) for MD and 1.5 dB (95% CI, 1.2-1.9 dB) for PSD. This was significantly better than mean thickness estimates for 10-2 MD (0.61 [95% CI, 0.47-0.71] and 3.0 dB [95% CI, 2.5-3.5 dB]) and 10-2 PSD (0.46 [95% CI, 0.31-0.60] and 2.3 dB [95% CI, 1.8-2.7 dB]). Combined DL models estimating 24-2 achieved R2 of 0.79 (95% CI, 0.72-0.84) for MD and 0.68 (95% CI, 0.53-0.79) for PSD and MAEs of 2.1 dB (95% CI, 1.8-2.5 dB) for MD and 1.5 dB (95% CI, 1.3-1.9 dB) for PSD. This was significantly better than mean thickness estimates for 24-2 MD (0.41 [95% CI, 0.26-0.57] and 3.4 dB [95% CI, 2.7-4.5 dB]) and 24-2 PSD (0.38 [95% CI, 0.20-0.57] and 2.4 dB [95% CI, 2.0-2.8 dB]). The GCIPL (R2 = 0.79) and GCC (R2 = 0.75) had the highest performance estimating 10-2 and 24-2 MD, respectively. Deep learning models improved estimates of functional loss from SD OCT imaging. Accurate estimates can help clinicians to individualize VF testing to patients.

Christopher M ，Bowd C ，Proudfoot JA ，Belghith A ，Goldbaum MH ，Rezapour J ，Fazio MA ，Girkin CA ，De Moraes G ，Liebmann JM ，Weinreb RN ，Zangwill LM ... -  《-》

Prediction of Visual Field Progression with Baseline and Longitudinal Structural Measurements Using Deep Learning.

Identifying glaucoma patients at high risk of progression based on widely available structural data is an unmet task in clinical practice. We test the hypothesis that baseline or serial structural measures can predict visual field (VF) progression with deep learning (DL). Development of a DL algorithm to predict VF progression. 3,079 eyes (1,765 patients) with various types of glaucoma and ≥5 VFs, and ≥3 years of follow-up from a tertiary academic center were included. Serial VF mean deviation (MD) rates of change were estimated with linear-regression. VF progression was defined as negative MD slope with p<0.05. A Siamese Neural Network with ResNet-152 backbone pre-trained on ImageNet was designed to predict VF progression using serial optic-disc photographs (ODP), and baseline retinal nerve fiber layer (RNFL) thickness. We tested the model on a separate dataset (427 eyes) with RNFL data from different OCT. The Main Outcome Measure was Area under ROC curve (AUC). Baseline average (SD) MD was 3.4 (4.9)dB. VF progression was detected in 900 eyes (29%). AUC (95% CI) for model incorporating baseline ODP and RNFL thickness was 0.813 (0.757-0.869). After adding the second and third ODPs, AUC increased to 0.860 and 0.894, respectively (p<0.027). This model also had highest AUC (0.911) for predicting fast progression (MD rate <1.0 dB/year). Model's performance was similar when applied to second dataset using RNFL data from another OCT device (AUC=0.893; 0.837-0.948). DL model predicted VF progression with clinically relevant accuracy using baseline RNFL thickness and serial ODPs and can be implemented as a clinical tool after further validation.

Mohammadzadeh V ，Wu S ，Besharati S ，Davis T ，Vepa A ，Morales E ，Edalati K ，Rafiee M ，Martinyan A ，Zhang D ，Scalzo F ，Caprioli J ，Nouri-Mahdavi K ... -  《-》

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 《-》