Detecting Diabetic Retinopathy from Retinal Images Using CUDA Deep Neural Network

Parmar, Ricky; Lakshmanan, L.; Swarnalatha, P.; Soundrapandiyan, Rajkumar

doi:10.1002/9781119439004.ch17

Cited by 6 publications

(4 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A complete overview of these architectures is found in [274]. There are several DL-based architectures related to DR detection [275][276][277][278][279][280][281][282][283][284][285]. Based on the clinical significance of automated DR detection, we group application of DL into three broad classes: (i) vessels segmentation, (ii)…”

Section: Overview Of Deep Learningmentioning

confidence: 99%

See 1 more Smart Citation

A comprehensive survey on computer-aided diagnostic systems in diabetic retinopathy screening

Tavakoli¹,

Kelley²

2021

Photo Acoustic and Optical Coherence Tomography Imaging, Volume 3

View full text Add to dashboard Cite

Diabetes Mellitus (DM) can lead to significant microvasculature disruptions that eventually causes diabetic retinopathy (DR), or complications in the eye due to diabetes. If left unchecked, this disease can increase over time and eventually cause complete vision loss. The general method to detect such optical developments is through examining the vessels, optic nerve head, microaneurysms, haemorrhage, exudates, etc. from retinal images. Ultimately this is limited by the number of experienced ophthalmologists and the vastly growing number of DM cases. To enable earlier and efficient DR diagnosis, the field of ophthalmology requires robust computer aided diagnosis (CAD) systems. Our review is intended for anyone, from student to established researcher, who wants to understand what can be accomplished with CAD systems and their algorithms to modeling and where the field of retinal image processing in computer vision and pattern recognition is headed. For someone just getting started, we place a special emphasis on the logic, strengths and shortcomings of different databases and algorithms frameworks with a focus on very recent approaches.

show abstract

Section: Overview Of Deep Learningmentioning

confidence: 99%

“…Many DL-based approaches have been introduced for detection and classification of different types of DR lesions such as MAs, and HEs [2,6,16,112,115,275,276,[297][298][299][300][301].…”

Section: Microaneurysms and Hemorrhages Detectionmentioning

confidence: 99%

A comprehensive survey on computer-aided diagnostic systems in diabetic retinopathy screening

Tavakoli¹,

Kelley²

2021

Photo Acoustic and Optical Coherence Tomography Imaging, Volume 3

View full text Add to dashboard Cite

show abstract

“…DNN have shown good performance diagnosing retinopathy from retinal fundus images, including datasets from Otago and Messidor [24], and three clinical departments in Sichuan Provincial People's Hospital [25]. Parmar et al [26] employed a convolutional neural network to detect DR from retinal images and their model outperformed others considered. Furthermore, the ResNet architecture model was utilized to detect DR from fundus images achieving an excellent classification accuracy [27,28].…”

Section: Deep Neural Networkmentioning

confidence: 99%

Deep Neural Network for Predicting Diabetic Retinopathy from Risk Factors

et al. 2020

View full text Add to dashboard Cite

Extracting information from individual risk factors provides an effective way to identify diabetes risk and associated complications, such as retinopathy, at an early stage. Deep learning and machine learning algorithms are being utilized to extract information from individual risk factors to improve early-stage diagnosis. This study proposes a deep neural network (DNN) combined with recursive feature elimination (RFE) to provide early prediction of diabetic retinopathy (DR) based on individual risk factors. The proposed model uses RFE to remove irrelevant features and DNN to classify the diseases. A publicly available dataset was utilized to predict DR during initial stages, for the proposed and several current best-practice models. The proposed model achieved 82.033% prediction accuracy, which was a significantly better performance than the current models. Thus, important risk factors for retinopathy can be successfully extracted using RFE. In addition, to evaluate the proposed prediction model robustness and generalization, we compared it with other machine learning models and datasets (nephropathy and hypertension–diabetes). The proposed prediction model will help improve early-stage retinopathy diagnosis based on individual risk factors.

show abstract

“…Pramar and Lakshmanan 2019 [15] proposed a new method to detect MAs and Exudates using Convolutional Neural Networks to extract lesion candidates and then for prediction of the model, they trained the model on a GPU. GPU-accelerated library of primitives aimed at Deep Neural Networks, NVIDIA CUDA Deep Neural Network (cuDNN) is used in their model.…”

Section: Introductionmentioning

confidence: 99%

Automated Approach for Extraction of Microaneurysms and Hemorrhages in Retinal Fundus Images

Jaafar¹,

Al-Libawy²,

Al-Gayem³

2020

IJIES

View full text Add to dashboard Cite

Early detection and quantification of microaneurysms and hemorrhages can potentially help reduce the risk of vision loss. In this paper, an automatic algorithm for extraction of microaneurysms and hemorrhages from color fundus images is presented. The color retinal image is analyzed and segmented using a top-down segmentation after some image enhancement steps, namely shade correction and contrast enhancement. The whole segmented images create a dataset of regions. Edge detection followed by region-growing techniques are also used to refine the top-down segmentation results. Segmented regions could be classified into true lesions and artifacts using a set of prior features like size, edge strength, color and texture. These are extracted and then utilized with a local-region properties classifier. To validate the outcomes of the proposed method, clinician's references were used to calculate the average performance measures. The proposed method achieved sensitivity of 98.8%, specificity of 97.7%, accuracy of 99.3%, and positive predictive value of 83.7% per-pixel basis. Superior performance measures and high computation speed have assured that the proposed method is more efficient and reproducible compared to the manual methods. The significant novelty of this work is in the use of three refining operations, i.e. Edge detection, region growing and classification to achieve optimal performance especially for the sensitivity and positive predictive value. The performance measures of the proposed algorithm are compared against many recent automated systems and are found to outperform most of them. The proposed algorithm performance is very close to that of the specialist ground truth. Hence, premium performance and high computation speed of this algorithm make it promising approach for a computer-aided mass screening of retinal diseases.

show abstract

Detecting Diabetic Retinopathy from Retinal Images Using CUDA Deep Neural Network

Cited by 6 publications

References 22 publications

A comprehensive survey on computer-aided diagnostic systems in diabetic retinopathy screening

A comprehensive survey on computer-aided diagnostic systems in diabetic retinopathy screening

Deep Neural Network for Predicting Diabetic Retinopathy from Risk Factors

Automated Approach for Extraction of Microaneurysms and Hemorrhages in Retinal Fundus Images

Contact Info

Product

Resources

About