A Prediction Method of Visual Field Sensitivity Using Fundus Autofluorescence Images in Patients With Retinitis Pigmentosa

Estimation of Visual Function Using Deep Learning From Ultra-Widefield Fundus Images of Eyes With Retinitis Pigmentosa

Importance

There is no widespread effective treatment to halt the progression of retinitis pigmentosa. Consequently, adequate assessment and estimation of residual visual function are important clinically.

Objective

To examine whether deep learning can accurately estimate the visual function of patients with retinitis pigmentosa by using ultra-widefield fundus images obtained on concurrent visits.

Design, Setting, and Participants

Data for this multicenter, retrospective, cross-sectional study were collected between January 1, 2012, and December 31, 2018. This study included 695 consecutive patients with retinitis pigmentosa who were examined at 5 institutions. Each of the 3 types of input images-ultra-widefield pseudocolor images, ultra-widefield fundus autofluorescence images, and both ultra-widefield pseudocolor and fundus autofluorescence images-was paired with 1 of the 31 types of ensemble models constructed from 5 deep learning models (Visual Geometry Group-16, Residual Network-50, InceptionV3, DenseNet121, and EfficientNetB0). We used 848, 212, and 214 images for the training, validation, and testing data, respectively. All data from 1 institution were used for the independent testing data. Data analysis was performed from June 7, 2021, to December 5, 2022.

Main Outcomes and Measures

The mean deviation on the Humphrey field analyzer, central retinal sensitivity, and best-corrected visual acuity were estimated. The image type-ensemble model combination that yielded the smallest mean absolute error was defined as the model with the best estimation accuracy. After removal of the bias of including both eyes with the generalized linear mixed model, correlations between the actual values of the testing data and the estimated values by the best accuracy model were examined by calculating standardized regression coefficients and P values.

Results

The study included 1274 eyes of 695 patients. A total of 385 patients were female (55.4%), and the mean (SD) age was 53.9 (17.2) years. Among the 3 types of images, the model using ultra-widefield fundus autofluorescence images alone provided the best estimation accuracy for mean deviation, central sensitivity, and visual acuity. Standardized regression coefficients were 0.684 (95% CI, 0.567-0.802) for the mean deviation estimation, 0.697 (95% CI, 0.590-0.804) for the central sensitivity estimation, and 0.309 (95% CI, 0.187-0.430) for the visual acuity estimation (all P < .001).

Conclusions and Relevance

Results of this study suggest that the visual function estimation in patients with retinitis pigmentosa from ultra-widefield fundus autofluorescence images using deep learning might help assess disease progression objectively. Findings also suggest that deep learning models might monitor the progression of retinitis pigmentosa efficiently during follow-up.

Association between the number of visual fields and the accuracy of future prediction in eyes with retinitis pigmentosa

Purpose

To evaluate the minimum number of visual fields (VFs) required to precisely predict future VFs in eyes with retinitis pigmentosa (RP).

Methods

A series of 12 VFs (Humphrey Field Analyzer 10–2 test (8.9 years in average) were analysed from 102 eyes of 52 patients with RP. The absolute error to predict the 12th VF using the prior 11 VFs was calculated in a pointwise manner, using the linear regression, and the 95% CI range was determined. Then, using 3–10 initial VFs, next VFs (4th to 11th VFs, respectively) were also predicted. The minimum number of VFs required for the mean absolute prediction error to reach the 95% CI was identified. Similar analyses were iterated for the second and third next VF predictions. Similar analyses were conducted using mean deviation (MD).

Results

In the pointwise analysis, the minimum number of VFs required to reach the 95% CI for the 12th VF was five (first and second next VF predictions) and six (third next VF prediction). For the MD analysis, three (first and second next VF predictions) and four (third next VF prediction) VFs were required to reach 95% CI for the 12th VF.

Conclusions

The minimum number of VFs required to obtain accurate predictions of the future VF was five or six in the pointwise analysis and three or four in the analysis with MD.

Detecting Progression of Retinitis Pigmentosa Using the Binomial Pointwise Linear Regression Method

Purpose

A method of evaluating central visual field (VF) progression in eyes with retinitis pigmentosa (RP) has still to be established. We previously reported the potential merit of applying a binomial test to pointwise linear regression (binomial PLR) in glaucoma progression. In the current study, we investigated the usefulness of binomial PLR in eyes with RP.

Methods

A series of 10 VFs (VF 1–10, Humphrey field analyzer, 10-2 test) from 196 eyes of 103 patients with RP were collected retrospectively. The PLR was performed by regressing the total deviation of all test points with the complete series of 10 VFs. The accuracy (positive predictive value, negative predictive value, and false-positive rate) and the time required to detect VF progression with shorter VF series (from VF 1–5 to VF 1–9) were compared across the binomial PLR, a permutation analysis of PLR (PoPLR), and a mean deviation (MD) trend analysis.

Results

In evaluating VF progression, the binomial PLR was comparable with the PoPLR and MD trend analyses in its positive predictive value (0.55 to 0.95), negative predictive value (0.67 to 0.92), and false-positive rate (0.01 to 0.05). The binomial PLR required significantly less time to detect VF progression (5.0 ± 2.0 years) than the PoPLR and MD trend analyses (P < 0.01, P < 0.001, respectively).

Conclusions

The application of a binomial PLR achieved reliable and earlier detection of central VF progression in eyes with RP.

Translational Relevance

A binomial PLR was useful in assessing VF progression in RP.

Development and validation of a visual field cluster in retinitis pigmentosa

The aim was to establish and evaluate a new clustering method for visual field (VF) test points to predict future VF in retinitis pigmentosa. A Humphrey Field Analyzer 10-2 test was clustered using total deviation values from 858 VFs. We stratified 68 test points into 24 sectors. Then, mean absolute error (MAE) of the sector-wise regression with them (S1) was evaluated using 196 eyes with 10 VF sequences and compared to pointwise linear regression (PLR), mean sensitivity of total area (MS) and also another sector-wise regression basing on VF mapping for glaucoma (29 sectors; S2). MAE with S1 were smaller than with PLR when between the first-third and first-seventh VFs were used. MAE with the method were significantly smaller than those of S2 when between the first-sixth and first-ninth VFs were used. The MAE of MS was smaller than those with S1 only when first to 3rd and first to 4th VFs were used; however, the prediction accuracy became far larger than any other methods when larger number of VFs were used. More accurate prediction was achieved using this new sector-wise regression than with PLR. In addition, the obtained cluster was more useful than that for glaucoma to predict progression.