Diabetic retinopathy (DR) is a disease that forms as a complication of diabetes. It is particularly dangerous since it often goes unnoticed and can lead to blindness if not detected early. Despite the clear importance and urgency of such an illness, there is no precise system for the early detection of DR so far. Fortunately, such system could be achieved using deep learning including convolutional neural networks (CNNs), which gained momentum in the field of medical imaging due to its capability of being effectively integrated into various systems in a manner that significantly improves the performance. This paper proposes a computer aided diagnostic (CAD) system for the early detection of non-proliferative DR (NPDR) using CNNs. The proposed system is developed for the optical coherence tomography (OCT) imaging modality. Throughout this paper, all aspects of deployment of the proposed system are studied starting from the preprocessing stage required to extract input retina patches to train the CNN without resizing the image, to the use of transfer learning principals and how to effectively combine features in order to optimize performance. This is done through investigating several scenarios for the system setup and then selecting the best one, which from the results revealed to be a two pre-trained CNNs based system, in which one of these CNNs is independently fed by nasal retina patches and the other one by temporal retina patches. The proposed transfer learning based CAD system achieves a promising accuracy of 94%. INDEX TERMS Convolutional neural network (CNN), diabetic retinopathy (DR), optical coherence tomography (OCT).
Diabetic retinopathy (DR) is a disease that forms as a complication of diabetes, It is particularly dangerous since it often goes unnoticed and can lead to blindness if not detected early. Despite the clear importance and urgency of such an illness, there is no precise system for the early detection of DR so far. Fortunately, such system could be achieved using deep learning including convolutional neural networks (CNNs), which gained momentum in the field of medical imaging due to its capability of being effectively integrated into various systems in a manner that significantly improves the performance. This paper proposes a computer aided diagnostic (CAD) system for the early detection of non-proliferative DR (NPDR) using CNNs. The proposed system is developed for the optical coherence tomography (OCT) imaging modality. Throughout this paper, all aspects of deployment of the proposed system are studied starting from the preprocessing stage required to extract input data to train the CNN without resizing the image, to the use of transfer learning principals and how best to combine features in order to optimize performance. A novel patch extraction framework for preprocessing is presented, followed by fovea detection algorithm, in addition to investigating the various CNN parameters for optimal deployment. Optimum CNN parameters and promising results are achieved. To the best of our knowledge, this is the first CNN-based DR early detection CAD system for OCT images. It achieves a promising accuracy of 94% with transfer learning. 4 recently emerged in the medical field, there is a continuous flux of interest in the 5 development of such systems due to their capability of improving the medical services 6 provided to the community in terms of accuracy and reliability in the diagnosis of 7 diseases. Meanwhile, machine learning is paving the way for breakthroughs in the 8 PLOS1/14 different areas of medical imaging such as in classification [1], segmentation [2], disease 9 detection [3], and image registration [4]. The application of deep learning, a subset of 10 machine learning algorithms, has made tremendous impact in the area of medical image 11 processing research [5, 6]. Deep learning is the latest emerging machine learning lead 12 technology in computer vision and image processing domains, particularly convolutional 13 neural networks (CNNs) [7]. They are especially powerful in solving problems that are 14 computationally difficult or with a high error rate such as medical image recognition 15with outstanding performance results [8]. For this reason, we got inspired to use CNNs 16 for the early detection of one of the most serious ophthalmological problems, which is 17 the diabetic retinopathy (DR).
18Blindness resulting from diabetes is turning to be an increasingly alarming issue, 19 which is due to the associated eye disease: DR. Such disease which develops as a 20 complication of diabetes, particularly type II [9, 10] occurs specifically from the chronic 21 high levels of sugar in the blood associated with ...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.