Improved Automated Detection of Diabetic Retinopathy on a Publicly Available Dataset Through Integration of Deep Learning.

By Michael David Abràmoff, Y. Lou, Ali Erginay, Warren Clarida, R. Amelon, J. Folk, Meindert Niemeijer. 2016

1,032 citations recorded by the source corpus.

Purpose To compare performance of a deep-learning enhanced algorithm for automated detection of diabetic retinopathy (DR), to the previously published performance of that algorithm, the Iowa Detection Program (IDP)-without deep learning components-on the same publicly available set of fundus images and previously reported consensus reference standard set, by three US Board certified retinal specialists. Methods We used the previously reported consensus reference standard of referable DR (rDR), defined as International Clinical Classification of Diabetic Retinopathy moderate, severe nonproliferative (NPDR), proliferative DR, and/or macular edema (ME). Neither Messidor-2 images, nor the three retinal specialists setting the Messidor-2 reference standard were used for training IDx-DR version X2.1. Sensitivity, specificity, negative predictive value, area under the curve (AUC), and their confidence intervals (CIs) were calculated. Results Sensitivity was 96.8% (95% CI: 93.3%-98.8%), specificity was 87.0% (95% CI: 84.2%-89.4%), with 6/874 false negatives, resulting in a negative predictive value of 99.0% (95% CI: 97.8%-99.6%). No cases of severe NPDR, PDR, or ME were missed. The AUC was 0.980 (95% CI: 0.968-0.992). Sensitivity was not statistically different from published IDP sensitivity, which had a CI of 94.4% to 99.3%, but specificity was significantly better than the published IDP specificity CI of 55.7% to 63.0%. Conclusions A deep-learning enhanced algorithm for the automated detection of DR, achieves significantly better performance than a previously reported, otherwise essentially identical, algorithm that does not employ deep learning. Deep learning enhanced algorithms have the potential to improve the efficiency of DR screening, and thereby to prevent visual loss and blindness from this devastating disease.

Subfields: Computer Vision and Pattern Recognition, Ophthalmology, Radiology, Nuclear Medicine and Imaging

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Ophthalmology

In Radiology, Nuclear Medicine and Imaging

Citations in the corpus

Contrastive Learning of Medical Visual Representations from Paired Images and Text (2020, methodology): “The first approach has been using high-quality annotations created by medical experts (Abràmoff et al., 2016; Gulshan et al., 2016; Shih et al., 2019; Wang and Wong, 2020).”
Zoom-in-Net: Deep Mining Lesions for Diabetic Retinopathy Detection (2017, methodology): “Recently, CNN based methods [4,3,9] have dramatically improved the performance of DR detection.”
A deep learning system for detecting diabetic retinopathy across the disease spectrum (2021, methodology): “Several deep learning algorithms with high speci ﬁ city and sensitivity have been developed for the classi ﬁ cation or detection of certain disease conditions based on medical images, including retinal images 19 – 23 .”
IDRiD: Diabetic Retinopathy - Segmentation and Grading Challenge. (2020, background): “There are some hybrid algorithms, in which multiple, semi-dependent CNN’s are trained based on the appearance of retinal lesions ( Abràmoff et al., 2016; Quellec et al., 2016 ).”
Artificial intelligence in retina. (2018, methodology): “…the sensitivity varied from 87% to 97%, and the specificity from 59% to 98% (97%/59% (Abramoff et al., 2013), use of 1748 images; 97%/87% (Abramoff et al., 2016), use of 1748 images; 87% (Gulshan et al., 2016), use of 1748 images; 91–100%/91%–92% (Ting et al., 2017), use of 76,370…”
Transfusion: Understanding Transfer Learning for Medical Imaging (2019, methodology): “, 2017) and ophthalmology, training Inception-v3, ResNet on retinal fundus images (Abràmo et al., 2016; Gulshan et al., 2016; Raghu et al., 2018; De Fauw et al., 2018).”
Deep Learning Applications in Medical Image Analysis (2018, methodology): “[57] evaluated a commercial device, the IDx-DR version X2.”