Deep Learning for Predicting Refractive Error From Retinal Fundus Images

Abstract

PURPOSE. We evaluate how deep learning can be applied to extract novel information such as refractive error from retinal fundus imaging. METHODS.Retinal fundus images used in this study were 45-and 30-degree field of view images from the UK Biobank and Age-Related Eye Disease Study (AREDS) clinical trials, respectively. Refractive error was measured by autorefraction in UK Biobank and subjective refraction in AREDS. We trained a deep learning algorithm to predict refractive error from a total of 226,870 images and validated it on 24,007 UK Biobank and 15,750 AREDS images. Our model used the ''attention'' method to identify features that are correlated with refractive error. RESULTS.The resulting algorithm had a mean absolute error (MAE) of 0.56 diopters (95% confidence interval [CI]: 0.55-0.56) for estimating spherical equivalent on the UK Biobank data set and 0.91 diopters (95% CI: 0.89-0.93) for the AREDS data set. The baseline expected MAE (obtained by simply predicting the mean of this population) was 1.81 diopters (95% CI: 1.79-1.84) for UK Biobank and 1.63 (95% CI: 1.60-1.67) for AREDS. Attention maps suggested that the foveal region was one of the most important areas used by the algorithm to make this prediction, though other regions also contribute to the prediction.CONCLUSIONS. To our knowledge, the ability to estimate refractive error with high accuracy from retinal fundus photos has not been previously known and demonstrates that deep learning can be applied to make novel predictions from medical images.

Predicting optical coherence tomography-derived diabetic macular edema grades from fundus photographs using deep learning

“…Our previous work has shown that deep learning can be leveraged to make predictions from fundus photographs, such as cardiovascular risk factors and refractive error, which are not possible by human experts 23,24 . This study describes a model that far exceeds expert performance for such a prediction, but one that has high clinical relevance and potentially important implications for screening programs worldwide.…”

“…A potential solution lies in the use of deep-learning algorithms, which have been applied to a variety of medical image classification tasks [14][15][16][17][18] , including for retinal imaging [19][20][21][22] . Encouragingly, in addition to achieving expert-level performance for grading fundus images, deep-learning algorithms are able to make predictions for which the underlying association with fundus images were previously unknown, such as cardiovascular risk factors 23 and refractive error 24 .…”

<p>Artificial Intelligence to Identify Retinal Fundus Images, Quality Validation, Laterality Evaluation, Macular Degeneration, and Suspected Glaucoma</p>

“…We believe, however, that the use of quality algorithms may be a more appropriate measure: it has a short feedback circle, even during the image taking process, thus offering a greater guarantee of quality and more cost-effectiveness. Other authors have also used quality algorithms, as was the case with Varadarajan et al, 20 who had to exclude 12% of the images of one of their datasets due to poor quality.…”

“…Dataset quality assessment is normally taken for granted, with researchers using trainings and certifications for photographers 17,18 or manually "cleaning" the datasets; 19 some authors have even used quality filters on algorithms before training the algorithms. 20 Optretina is a telemedicine platform which performs general screening for retinal diseases using nonmydriatic cameras and human evaluation by a retinal specialist ophthalmologist. 21 Our research in artificial intelligence deals with both data acquisition and the diagnostic steps of our telemedicine platform, in a near future we expect AI could assist our specialists in achieving levels of consistency and accuracy beyond unassisted human abilities.…”

Vaskuläre Biomarker der retinalen Gefäßanalyse

“…Dazu gehören bspw. die computergestützte Fluoreszenzangiografie [9,10] und die moderne OCT-Angiografie [11,12], die Vermessung von retinaler Gefäßwanddicke sowie longitudinaler Gefäßwandstruktur mittels konfokaler Mikroskopie [13,14] und der Einsatz des maschinellen Lernens bei Fundusfotografien zur Prognose des kardiovaskulären Risikos [15,16].…”