Dataset Information

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

ABSTRACT:

Purpose

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.

Design

Evaluation of a diagnostic technology.

Participants

A total of 9765 visual field (VF) SD OCT pairs collected from 1194 participants with and without GVFD (1909 eyes).

Methods

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.

Main outcome measures

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).

Results

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.

Conclusions

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.

SUBMITTER: Christopher M

PROVIDER: S-EPMC8063221 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:PURPOSE:To compare the diagnostic ability of wide-field swept-source optical coherence tomography (SS-OCT) retinal nerve fiber layer (RNFL) maps with spectral-domain OCT (SD-OCT) maps for detection of preperimetric (PPG) and early glaucoma (EG). PATIENTS AND METHODS:One hundred and forty-six eyes, including 37 healthy eyes, 38 eyes with PPG, and 71 eyes with EG, were analyzed. The patients underwent both SD-OCT (Cirrus HD-OCT; Carl Zeiss Meditec, Dublin, CA, USA) and wide-field SS-OCT scanning (DRI-OCT-1 Atlantis; Topcon, Tokyo, Japan). By SD-OCT, circumpapillary RNFL and macular ganglion cell analyses were performed. SS-OCT provides a wide-field RNFL thickness map and a SuperPixel map, which are composed of an RNFL deviation map of the peripapillary area and a deviation map of the composition of the ganglion cell layer with the inner plexiform layer and RNFL [GC-IPL+RNFL] in the macular area. The ability to discriminate PPG and EG from healthy eyes was assessed according to sensitivity, specificity and area under the receiver operating characteristic curve for parameters and criteria provided by SD-OCT and wide-field SS-OCT scanning. RESULTS:The wide-field RNFL thickness map obtained by SS-OCT showed the highest sensitivity to PPG and EG (92.1 and 97.2%, respectively) as compared with the other, SD-OCT criteria. The wide-field RNFL thickness map showed PPG-diagnostic performance comparable to the SD-OCT RNFL thickness and GC-IPL deviation maps (p?=?0.453 and 0.180), and PPG-diagnostic performance superior to the SD-OCT RNFL deviation and GC-IPL thickness maps (p?=?0.003 and 0.039). In EG, the wide-field RNFL thickness and SuperPixel maps showed diagnostic performance comparable to the SD-OCT thickness and deviation maps (p?=?0.065 to 0.100), except for the GC-IPL thickness map (p?=?0.004). CONCLUSIONS:The wide-field SS-OCT RNFL thickness maps showed a diagnostic ability for distinguishing PPG and EG from healthy eyes that was similar to that of SD-OCT. In the clinical setting, these maps can be effective for detection of early-glaucomatous changes.