Stroke diagnosis from retinal fundus images using multi texture analysis

Jeena, R. S.; Kumar, Amod; Mahadevan, Kittappa M.

doi:10.3233/jifs-169914

Cited by 14 publications

(3 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…17 A new architecture for segmentation process is introduce in Oliveira et al, 4 using a multi-scale fully convolutional neural network and stationary wavelet transform which generate the more specific results for the extraction of blood vessels. [18][19][20] Introduce various algorithms related to image processing like they use texture features, wavelet transform and support vector machine to control nonlinear nature of the data of images and to produce segmentation architecture for the vital contrast enhanced images. 21,22 Miri and Mahloojifar 23 applied curvelet transform in hybrid with deep leaning method to magnify the features of data, that is, enhance the edges of images.…”

Section: Related Workmentioning

confidence: 99%

“…For the training and testing with deep CNN models, the image patches are used because of the limited number of images in datasets and the need to reduce computational costs 17 . A new architecture for segmentation process is introduce in Oliveira et al, 4 using a multi‐scale fully convolutional neural network and stationary wavelet transform which generate the more specific results for the extraction of blood vessels 18–20 . Introduce various algorithms related to image processing like they use texture features, wavelet transform and support vector machine to control nonlinear nature of the data of images and to produce segmentation architecture for the vital contrast enhanced images 21,22 .…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

A deep data‐driven approach for enhanced segmentation of blood vessel for diabetic retinopathy

Yadav

2022

Int J Imaging Syst Tech

View full text Add to dashboard Cite

The segmentation step of retinal blood vessel helps to diagnosis the diseases including diabetic retinopathy, glaucoma, etc. The automatic image segmentation process helps experts to speed up the diagnosis of DR, since analytic methods are time consuming and error prone. The neural network (NN) based methods like U‐Net uses leap bonding that extract fine information from the training dataset. However automatic segmentation of image using neural network is a challenging process because of uneven and irregular geometry of organ. In this article, we proposed a U‐Net based approach for segmentation of retinal vessels. Before applying segmentation step, the affected area of image is enhanced with some preprocessing techniques. Then a dual tree discrete Ridgelet transform (DT‐DRT) is apply on the dataset to extract the features from the region of interest. The features accumulation with DT‐DRT ensures better feature representation of vessel for segmentation task. The proposed segmentation is implemented on different publicly available dataset and achieve accuracy of 96.01% in CHASE DB1, 97.65% in DRIVE and 98.61% in STARE dataset. The performance of this algorithm is also compared with some other deep learning models, and results demonstrate that this proposed algorithm performed better than them.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

A deep data‐driven approach for enhanced segmentation of blood vessel for diabetic retinopathy

Yadav

2022

Int J Imaging Syst Tech

View full text Add to dashboard Cite

show abstract

“…The proposed method is evaluated on PROSTATEx-2 2017 challenge dataset and demonstrated better results than the challenge winning method. In [9], the authors employ multi-texture descriptions in order to assist in predicting cardiovascular diseases such as stroke from retinal fundus images, i.e. images of the back of the eye.…”

mentioning

confidence: 99%