Original articleAutomated Identification of Diabetic Retinopathy Using Deep Learning
Section snippets
Methods
Figure 1A represents an abstraction of our algorithmic pipeline. We compiled and preprocessed fundus images across various sources into a large-scale data set. Our deep learning network learned data-driven features from this data set, characterizing DR based on an expert-labelled ground truth. These deep features were propagated (along with relevant metadata) into a tree-based classification model that output a final, actionable diagnosis.
Results
We tested the model using 5-fold stratified cross-validation on our local data set of 75 137 images, preserving the percentage of samples of each class per fold. This testing procedure trained 5 separate models, each holding out a distinct validation bucket of approximately 15 000 images. Average metrics were derived from 5 test runs on respective held-out data by comparing the model's predictions with the gold standard determined by the panel of specialists. A final, complete model was trained
Discussion
This study proposed a novel automated-feature learning approach to DR detection using deep learning methods. It provides a robust solution for DR detection within a large-scale data set, and the results attained indicate the high efficacy of our computer-aided model in providing efficient, low-cost, and objective DR diagnostics without depending on clinicians to examine and grade images manually. Our method also does not require any specialized, inaccessible, or costly computer equipment to
References (34)
- et al.
Proposed international clinical diabetic retinopathy and diabetic macular edema disease severity scales
Ophthalmology
(2003) - et al.
Global estimates of diabetes prevalence for 2013 and projections for 2035
Diabetes Res Clin Pract
(2014) - et al.
Computer-aided diagnosis of diabetic retinopathy: a review
Comput Biol Med
(2013) - et al.
Algorithms for digital image processing in diabetic retinopathy
Comput Med Imaging Graph
(2009) - et al.
Discrimination of retinal images containing bright lesions using sparse coded features and SVM
Comput Biol Med
(2015) - et al.
TeleOphta: machine learning and image processing methods for teleophthalmology
IRBM
(2013) - et al.
An ensemble-based system for automatic screening of diabetic retinopathy
Knowledge-Based Syst
(2014) IDF Diabetes Atlas 7th edition
(2015)- et al.
Global estimates of undiagnosed diabetes in adults
Diabetes Res Clin Pract
(2015) - et al.
Trends in health care expenditure in U.S. adults with diabetes: 2002–2011
Diabetes Care
(2015)
Grader agreement, and sensitivity and specificity of digital photography in a community optometry-based diabetic eye screening program
Clin Ophthalmol
Screening for diabetic retinopathy in rural area using single-field, digital fundus images
J Med Assoc Thail
Improved automated detection of diabetic retinopathy on a publicly available dataset through integration of deep learning
Invest Opthalmol Vis Sci
SURF descriptor and pattern recognition techniques in automatic identification of pathological retinas. 2015 Brazilian Conference on Intelligent Systems (BRACIS)
IEEE
Convolutional networks and applications in vision
Deep learning
Nature
Cited by (0)
Supplemental material available at www.aaojournal.org.
Financial Disclosure(s): The author(s) have made the following disclosure(s): R.G.: Patent - (Patent Application Number: 62383333); Patent Filing date: September 2, 2016.
Author Contributions:
Conception and design: Gargeya, Leng
Analysis and interpretation: Gargeya, Leng
Data collection: Gargeya, Leng
Obtained funding: none
Overall responsibility: Gargeya, Leng