Search

In recent years, several studies have undergone automatic blood vessel segmentation based on unsupervised and supervised algorithms to reduce user interruption. Deep learning networks have been used to get highly accurate segmentation results. However, the incorrect segmentation of pathological information and low micro-vascular segmentation is considered the challenges present in the existing methods for segmenting the retinal blood vessel. It also affects different degrees of vessel thickness, contextual feature fusion in technique, and perception of details. A deep learning-aided method has been presented to address these challenges in this paper. In the first phase, the preprocessing is performed using the retinal fundus images employed by the black ring removal, LAB conversion, CLAHE-based contrast enhancement, and grayscale image. Thus, the blood vessel segmentation is performed by a new deep learning model termed optimized ResUNet + +. As an improvement to this deep learning architecture, the activation function is optimized by the J-AGSO algorithm. The objective function for the optimized ResUNet + + -based blood vessel segmentation is to minimize the binary cross-entropy loss function. Further, the post-processing of the images is carried out by median filtering and binary thresholding. By verifying the standard benchmark datasets, the proposed model outperforms and attains enhanced performance. [ABSTRACT FROM AUTHOR]

Glaucoma is looked on as the most important cause of irremediable vision loss worldwide. Early detection of eye diseases, especially glaucoma is serious for preparing timely medical care and keep downing the vision loss. In this paper, a fast segmentation algorithm is proposed which is based on a new simplified U-Net architecture for optic disc and retinal vessels segmentation. The proposed method includes a modified and reinforced structure that will reduce the prediction time while maintaining the performance and accuracy at an comparable level due to other state of the art methods. For example, for optic disc segmentation, the proposed method can segment the optic disc in 0.008 seconds on DRIONS-DB dataset, and for vessels segmentation,it can segment in 0.03 seconds on DRIVE dataset. According to these results and an extension of the proposed method can be used as a real-time intelligent medical system which able to be implemented on the usual hardware equipement in the ophthalmology clinics. This method, which can perform optic disc and retinal vessels segmentation tasks in a short time, increases the performance of ophthalmologists in glaucoma diagnosing. [ABSTRACT FROM AUTHOR]

This paper proposes a fundus retinal blood vessel segmentation model based on a deep convolutional network structure and biological visual feature extraction mechanism. It aims to solve the multi-scale problem of blood vessels in the fundus retinal blood vessel segmentation task in the field of medical image processing on the basis of increasing the biological interpretability of the model. First, the subject feature information of the retinal blood vessel image is obtained by using the non-subsampled Residual Bolck convolution main channel. Secondly, combined with the study of biological vision mechanisms, an information processing model of the Retina-Exogenius-Primary visual cortex (V1) ventral visual pathway was established. Gabor functions of different scales are used to simulate the structure of different levels of the visual pathway, and the scale information at different levels is integrated into the corresponding hierarchical stages of the convolutional main pathway network to enrich the information of small blood vessels and enhance the semantic information of the overall blood vessels. Finally, considering the imbalance of the ratio of vessel and nonvessel pixels, an adaptive optimization scheme using hybrid loss function weights is proposed to enhance the priority of blood vessel pixels in the calculation of the loss function. According to the experimental results on the STARE, DRIVE and CHASE_DB1 data sets, the model still achieves superior performance evaluation indicators overall compared with the existing optimal methods in the fundus retinal blood vessel segmentation task. This research is of great significance to the field of medical image processing and can provide more accurate auxiliary diagnostic information for clinical diagnosis and treatment. [ABSTRACT FROM AUTHOR]

The segmentation of blood vessels through color fundus images is a difficult and time-consuming task that requires experienced clinicians. Recently, researchers have shown that blood vessel segmentation using methods based on deep neural networks has achieved highly satisfactory results. This motivates us to employ a fast and accurate deep learning-based method that can be used in blood vessel segmentation. Five improved deep learning-based networks (U-Net, DenseU-Net, LadderNet, R2U-Net, and ATTU-Net), along with an enhanced customized R2-ATT U-Net deep learning network, have been employed to segment the retinal blood vessel tree. Segmentation using patches extraction is executed, where we have used 10,000 patches per image, in total, 8000 for training and 2000 for testing public benchmark STARE dataset images. Initially, we performed the training, followed by testing for all six models using the patch extraction approach. All the aspects of the testing phase (test log, ROC curve, precision recall curves, and confusion matrix) and statistical performance measuring metrics (accuracy, sensitivity, specificity, F1 score, precision, and AUC values) are covered in this work and are also shown in the form of tables and graphs. An in-depth performance evaluation analysis of these six implemented improved nets has been performed to evaluate the segmentation process. We got the best result in the case of LadderNet, with an accuracy of 0.971, which is comparable to recent state-of-the-art studies. In terms of accuracy, the enhanced customized R2-ATT U-Net deep learning network is also highly satisfactory, being near the best model, LadderNet. The experiment results demonstrate that the proposed methodology achieves satisfactory performance in the retinal blood vessel extraction domain and can help ophthalmologists predict many eye-related diseases at a preliminary stage. [ABSTRACT FROM AUTHOR]