Blood vessel segmentation of retinal fundus images using dynamic preprocessing and mathematical morphology

Chakour, E.; Mrad, Y.; Mansouri, Ali; Elloumi, Yaroub; Bedoui, Sammy; Andaloussi, Idriss Benatiya; Ahaitouf, Ali

doi:10.1109/codit55151.2022.9804004

Cited by 7 publications

(2 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Retina image features such as spots, lesions, vessels can be extracted by first having the images preprocessed to create a clear view of the features before extraction. Mathematical morphology technique [47] can effectively be used to preprocess images and display these features as the image content for quality evaluation. To maximize feature extraction, an image can be divided into zones using a morphological image structuring element.…”

Section: Resultsmentioning

confidence: 99%

Retinal Image Quality Assessment Using Morphological Operations

Ndunge Mutua,

Shibwabo Kasamani,

Reich

2024

IJCDS

View full text Add to dashboard Cite

Retinopathy of Prematurity (ROP) disease affects newborn babies born preterm. The disease has five stages, with stage IV and V being critical where if the disease is not diagnosed at stage III when the vessels begin to grow abnormally, reversing it is not possible. Diagnosis and treatment are possible between stage I-III. Hospitals without eye specialists, a doctor can be instructed on how to capture retina image which is transmitted online to an ophthalmologist for disease diagnosis. Different devices produce images of varying qualities and during transmission, some image features could be lost. Some images are captured under poor lighting conditions resulting to poor quality images being generated. This study proposes an algorithm which performs quality assessment of retina images before being used to diagnose ROP Stage II or III disease. The algorithm was developed and tested using Retinopathy of Prematurity disease data of 91 images available at the Kaggle database and the objective was to separate images of quality from non-quality ones. The algorithm was able to separate quality from non-quality retina images with 92.82% sensitivity, 96.98% specificity and 97.31% accuracy. Performance evaluation was conducted by means of estimating the similarity measure of DSC and Jaccard index (JI), producing agreeable indices of 94.81% DSC and 88.42% JI.

show abstract

Section: Resultsmentioning

confidence: 99%

Retinal Image Quality Assessment Using Morphological Operations

Ndunge Mutua,

Shibwabo Kasamani,

Reich

2024

IJCDS

View full text Add to dashboard Cite

show abstract

“…Among the most popular are deep neural networks [4], particularly U-network-based architectures [5], directional convolutional kernels [6], various approaches to region growing [7] and applied mathematical morphology [8]. In recent years, there have been several proposals to improve the response of hessian-based methods by, for example, using swarm optimization [9], genetic programming and other approaches [10][11][12][13][14][15][16][17].…”

Section: Introductionmentioning

confidence: 99%

High-Level Hessian-Based Image Processing with the Frangi Neuron

Hachaj,

Piekarczyk

2023

Electronics

View full text Add to dashboard Cite

The Frangi neuron proposed in this work is a complex element that allows high-level Hessian-based image processing. Its adaptive parameters (weights) can be trained using a minimum number of training data. In our experiment, we showed that just one image is enough to optimize the values of the weights. An intuitive application of the Frangi neuron is to use it in image segmentation process. In order to test the performance of the Frangi neuron, we used diverse medical datasets on which second-order structures are visualized. The Frangi network presented in this paper trained on a single image proved to be significantly more effective than the U-net trained on the same dataset. For the datasets tested, the network performed better as measured by area under the curve receiver operating characteristic (ROC AUC) than U-net and the Frangi algorithm. However, the Frangi network performed several times faster than the non-GPU implementation of Frangi. There is nothing to prevent the Frangi neuron from being used as part of any other network as a component to process two-dimensional images, for example, to detect certain second-order features in them.

show abstract