Robust retinal blood vessel segmentation using hybrid active contour model

Karn, Prakash Kumar; Biswal, Birendra; Samantaray, Subhransu Ranjan

doi:10.1049/iet-ipr.2018.5413

Cited by 51 publications

(38 citation statements)

References 54 publications

(54 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The average Sp is almost at par with the first position orientation scores method by Zhang [34] at 0.9758. The average Acc measure of our method lies in third place close behind Karn [36] and Zhang [34] with Acc scores of 0.96 and 0.9554, respectively. Overall, our method consistently ranks in the top tier of all unsupervised vessel segmentation techniques for the STARE dataset.…”

Section: Resultsmentioning

confidence: 84%

“…The highest Sp score is 0.99 for image 4 while the best Acc is 0.97 for image 19. Talking about Sp, our method just takes the edge over Zhang [34] with 0.9725, but loses the first place to Karn [36] with a Sp of 0.98. The same hybrid active contour technique by Karn [36] scores the highest Acc of all methods (0.97) with our method in second place.…”

Section: Resultsmentioning

confidence: 99%

“…Talking about Sp, our method just takes the edge over Zhang [34] with 0.9725, but loses the first place to Karn [36] with a Sp of 0.98. The same hybrid active contour technique by Karn [36] scores the highest Acc of all methods (0.97) with our method in second place. It is pertinent to note here that Karn [36] achieved these numbers at the cost of the method’s Sn and also rounded-off his performance parameters in his report to a decimal place accuracy of 2.…”

Section: Resultsmentioning

confidence: 99%

“…A curvature analysis followed by adaptive local thresholding resulted in excellent all-round performance parameters on both DRIVE and STARE datasets. Karn et al [36] constructed a hybrid active contour model based on a Gradient Vector Flow (GVF) framework to focus on tiny blood vessels. A novel preprocessing technique coupled with phase-based binarization helped this method report high accuracy scores and effective vessel segmentation.…”

Section: Literature Reviewmentioning

confidence: 99%

See 3 more Smart Citations

A Multi-Scale Directional Line Detector for Retinal Vessel Segmentation

Khawaja

Khan

et al. 2019

Sensors

View full text Add to dashboard Cite

The assessment of transformations in the retinal vascular structure has a strong potential in indicating a wide range of underlying ocular pathologies. Correctly identifying the retinal vessel map is a crucial step in disease identification, severity progression assessment, and appropriate treatment. Marking the vessels manually by a human expert is a tedious and time-consuming task, thereby reinforcing the need for automated algorithms capable of quick segmentation of retinal features and any possible anomalies. Techniques based on unsupervised learning methods utilize vessel morphology to classify vessel pixels. This study proposes a directional multi-scale line detector technique for the segmentation of retinal vessels with the prime focus on the tiny vessels that are most difficult to segment out. Constructing a directional line-detector, and using it on images having only the features oriented along the detector’s direction, significantly improves the detection accuracy of the algorithm. The finishing step involves a binarization operation, which is again directional in nature, helps in achieving further performance improvements in terms of key performance indicators. The proposed method is observed to obtain a sensitivity of 0.8043, 0.8011, and 0.7974 for the Digital Retinal Images for Vessel Extraction (DRIVE), STructured Analysis of the Retina (STARE), and Child Heart And health Study in England (CHASE_DB1) datasets, respectively. These results, along with other performance enhancements demonstrated by the conducted experimental evaluation, establish the validity and applicability of directional multi-scale line detectors as a competitive framework for retinal image segmentation.

show abstract

Section: Resultsmentioning

confidence: 84%

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

See 2 more Smart Citations

A Multi-Scale Directional Line Detector for Retinal Vessel Segmentation

Khawaja

Khan

et al. 2019

Sensors

View full text Add to dashboard Cite

show abstract

“…According to the summarized retinal vessel segmentation algorithms [6,7], it can be divided into two categories: unsupervised algorithms and supervised algorithms. In general, unsupervised learning algorithms do not require manual annotation data and mainly use some pre-set rules to extract vessel features and achieve segmentation, such as matching filter-based algorithms [8], deformable modelbased algorithms [9], and tracking-based algorithms [10]. However, fixed segmentation rules often cannot match the diversity of vascular morphological distribution.…”

Section: Introductionmentioning

confidence: 99%

Retinal Vessel Segmentation Combined With Generative Adversarial Networks and Dense U-Net

Guo

Chen

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Retinal blood vessels are the basis for clinical diagnosis of some diseases. Achieving automatic retinal vessel segmentation from fundus images is an important and challenging work. In this paper, a neural network architecture based on the Dense U-net using Inception module is proposed for retinal vessel segmentation. First, the skip connections in traditional U-net are replaced with Dense Block to achieve full fusion of features from shallow layers to deep layers. Then, the Inception module is applied to supersede the traditional convolution operation. Thus, vessel features corresponding to the convolution kernels of different sizes can be extracted. Finally, Generative Adversarial Networks (GAN) are adopted in the training phase. The Dense U-net using Inception module is treated as the generator of GAN, and a multilayer neural network is created as the discriminator of GAN. The generator and discriminator are trained alternately. The loss function is a combination of segmentation loss and GAN loss. So that segmentation results can be fitted to the ground truth from both pixel value and pixel distribution. The algorithm proposed in this paper is verified on the public Digital Retinal Images for Vessel Extraction (DRIVE) dataset, where the Dice rate reaches 82.15%, and the AU-ROC and AU-PR reach 0.9772 and 0.9058, respectively. Experiments show that the proposed algorithm is effective in realizing automatic retinal vessel segmentation.

show abstract

Robust segmentation of optic disc and optic cup using statistical Kurtosis test

Biswal¹,

Eadara²,

Bebarta³

et al. 2019

Int J Imaging Syst Tech

View full text Add to dashboard Cite

Glaucoma is a chronic and irreversible eye disease that leads to the loss of vision. Evaluation of the Cup to Disc Ratio (CDR) plays a prominent role in the early detection of glaucoma. This paper presents a novel algorithm to compute the CDR for the fundus images. In order to calculate the CDR, the vertical diameter of Optic Disc (OD) and the vertical diameter of the Optic Cup (OC) are calculated from the segmented OD and segmented OC, respectively. This study presents OD and OC segmentation algorithms based on a new statistics‐based kurtosis test. A unique OC boundary segmentation method is presented, which is the combination of partial cup extraction and the cup boundary under the Central Retinal Blood vessels (CRBV). A novel preprocessing technique is introduced to extract the CRBV from the automated Region of Interest (ROI). The experimental results confirm that the proposed algorithm outperforms the state‐of‐art OD and OC segmentation on the three publicly available datasets: ORIGA, DIARETDB0, and DIARETDB1. The proposed OD and OC segmentation achieve accuracies of up to 0.99 and 0.97, respectively. In addition, the proposed model achieves excellent CDR evaluation with an average error percentage reduced to up to 9.6496 for the considered datasets.

show abstract

Robust retinal blood vessel segmentation using hybrid active contour model

Cited by 51 publications

References 54 publications

A Multi-Scale Directional Line Detector for Retinal Vessel Segmentation

A Multi-Scale Directional Line Detector for Retinal Vessel Segmentation

Retinal Vessel Segmentation Combined With Generative Adversarial Networks and Dense U-Net

Robust segmentation of optic disc and optic cup using statistical Kurtosis test

Contact Info

Product

Resources

About