scholarly journals Detection of Glaucoma in Retinal Fundus Images using Fast Fuzzy C Mean Clustering

2020 ◽  
Vol 7 ◽  
Keyword(s):  
Optic Disc ◽  
Vision Loss ◽  
Fluid Pressure ◽  
Fundus Images ◽  
New Approach ◽  
Optic Cup ◽  
Retinal Fundus Images ◽  
Retinal Fundus

Glaucoma is one of the major causes of vision loss in today’s world. Glaucoma is a disease in the eye where fluid pressure in the eye increases; if it is not timely cured, the patient may lose their vision. Glaucoma can be detected by examining boundary of optics cup and optics disc acquired from fundus images. The proposed method suggest automatic detect the boundary of optics cup and optics disc with processing of fundus images. This paper explores the new approach fast fuzzy C-mean technique for segmenting the optic disc and optic cup in fundus images. Results evaluated by fast fuzzy C mean a technique is faster than fuzzy C-mean method. The proposed method reported results to 91.91%, 90.49% and 90.17% when tested on DRIONS, DRIVE and STARE on publicly available databases of fundus images.

scholarly journals Multi-Path Recurrent U-Net Segmentation of Retinal Fundus Image

2020 ◽  
Vol 10 (11) ◽  
pp. 3777 ◽  
Author(s):  
Yun Jiang ◽  
Falin Wang ◽  
Jing Gao ◽  
Simin Cao
Keyword(s):  
Blood Vessel ◽  
Optic Disc ◽  
Vessel Segmentation ◽  
Fundus Images ◽  
Fundus Image ◽  
Retinal Blood Vessel ◽  
Blood Vessel Segmentation ◽  
Optic Cup ◽  
Retinal Fundus Images ◽  
Retinal Fundus

Diabetes can induce diseases including diabetic retinopathy, cataracts, glaucoma, etc. The blindness caused by these diseases is irreversible. Early analysis of retinal fundus images, including optic disc and optic cup detection and retinal blood vessel segmentation, can effectively identify these diseases. The existing methods lack sufficient discrimination power for the fundus image and are easily affected by pathological regions. This paper proposes a novel multi-path recurrent U-Net architecture to achieve the segmentation of retinal fundus images. The effectiveness of the proposed network structure was proved by two segmentation tasks: optic disc and optic cup segmentation and retinal vessel segmentation. Our method achieved state-of-the-art results in the segmentation of the Drishti-GS1 dataset. Regarding optic disc segmentation, the accuracy and Dice values reached 0.9967 and 0.9817, respectively; as regards optic cup segmentation, the accuracy and Dice values reached 0.9950 and 0.8921, respectively. Our proposed method was also verified on the retinal blood vessel segmentation dataset DRIVE and achieved a good accuracy rate.

Application of conditional GAN models in optic disc/optic cup segmentation of retinal fundus images

2021 ◽  
Author(s):  
Tales H. Carvalho ◽  
Carlos H. Moraes ◽  
Rafael C. Almeida ◽  
Danilo H. Spadoti
Keyword(s):  
Optic Disc ◽  
Fundus Images ◽  
Optic Cup ◽  
Retinal Fundus Images ◽  
Retinal Fundus

scholarly journals Two-Stage Mask-RCNN Approach for Detecting and Segmenting the Optic Nerve Head, Optic Disc, and Optic Cup in Fundus Images

2020 ◽  
Vol 10 (11) ◽  
pp. 3833 ◽  
Author(s):  
Haidar Almubarak ◽  
Yakoub Bazi ◽  
Naif Alajlan
Keyword(s):  
Optic Nerve ◽  
Optic Disc ◽  
Optic Nerve Head ◽  
Fine Tuning ◽  
Fundus Images ◽  
Two Stage ◽  
Second Stage ◽  
Retinal Fundus Images ◽  
Tuning Strategy ◽  
Retinal Fundus

In this paper, we propose a method for localizing the optic nerve head and segmenting the optic disc/cup in retinal fundus images. The approach is based on a simple two-stage Mask-RCNN compared to sophisticated methods that represent the state-of-the-art in the literature. In the first stage, we detect and crop around the optic nerve head then feed the cropped image as input for the second stage. The second stage network is trained using a weighted loss to produce the final segmentation. To further improve the detection in the first stage, we propose a new fine-tuning strategy by combining the cropping output of the first stage with the original training image to train a new detection network using different scales for the region proposal network anchors. We evaluate the method on Retinal Fundus Images for Glaucoma Analysis (REFUGE), Magrabi, and MESSIDOR datasets. We used the REFUGE training subset to train the models in the proposed method. Our method achieved 0.0430 mean absolute error in the vertical cup-to-disc ratio (MAE vCDR) on the REFUGE test set compared to 0.0414 obtained using complex and multiple ensemble networks methods. The models trained with the proposed method transfer well to datasets outside REFUGE, achieving a MAE vCDR of 0.0785 and 0.077 on MESSIDOR and Magrabi datasets, respectively, without being retrained. In terms of detection accuracy, the proposed new fine-tuning strategy improved the detection rate from 96.7% to 98.04% on MESSIDOR and from 93.6% to 100% on Magrabi datasets compared to the reported detection rates in the literature.

An Explainable Deep Learning Ensemble Model for Robust Diagnosis of Diabetic Retinopathy Grading

2021 ◽  
Vol 17 (3s) ◽  
pp. 1-24
Author(s):  
Mohammad Shorfuzzaman ◽  
M. Shamim Hossain ◽  
Abdulmotaleb El Saddik
Keyword(s):  
Diabetic Retinopathy ◽  
Deep Learning ◽  
Vision Loss ◽  
Final Diagnosis ◽  
Learning Rate ◽  
Ensemble Model ◽  
Fundus Images ◽  
Robust Detection ◽  
Retinal Fundus Images ◽  
Retinal Fundus

Diabetic retinopathy (DR) is one of the most common causes of vision loss in people who have diabetes for a prolonged period. Convolutional neural networks (CNNs) have become increasingly popular for computer-aided DR diagnosis using retinal fundus images. While these CNNs are highly reliable, their lack of sufficient explainability prevents them from being widely used in medical practice. In this article, we propose a novel explainable deep learning ensemble model where weights from different models are fused into a single model to extract salient features from various retinal lesions found on fundus images. The extracted features are then fed to a custom classifier for the final diagnosis of DR severity level. The model is trained on an APTOS dataset containing retinal fundus images of various DR grades using a cyclical learning rates strategy with an automatic learning rate finder for decaying the learning rate to improve model accuracy. We develop an explainability approach by leveraging gradient-weighted class activation mapping and shapely adaptive explanations to highlight the areas of fundus images that are most indicative of different DR stages. This allows ophthalmologists to view our model's decision in a way that they can understand. Evaluation results using three different datasets (APTOS, MESSIDOR, IDRiD) show the effectiveness of our model, achieving superior classification rates with a high degree of precision (0.970), sensitivity (0.980), and AUC (0.978). We believe that the proposed model, which jointly offers state-of-the-art diagnosis performance and explainability, will address the black-box nature of deep CNN models in robust detection of DR grading.

Sign in / Sign up

Export Citation Format

Share Document