Wyniki wyszukiwania - BazTech

1

Dual-channel asymmetric convolutional neural network for an efficient retinal blood vessel segmentation in eye fundus images

Xu Yanan, Fan Yingle

Biocybernetics and Biomedical Engineering

|

2022

|

Vol. 42, no. 2

695--706

EN

The morphological properties of retinal vessels are closely related to the diagnosis of ophthalmic diseases. However, many problems in retinal images, such as complicated directions of vessels and difficult recognition of capillaries, bring challenges to the accurate segmentation of retinal blood vessels. Thus, we propose a new retinal blood vessel segmentation method based on a dual-channel asymmetric convolutional neural network (CNN). First, we construct the thick and thin vessel extraction module based on the morphological differences in retinal vessels. A two-dimensional (2D) Gabor filter is used to perceive the scale characteristics of blood vessels after selecting the direction of blood vessels; thereby, adaptively extracting the thick vessel features characterizing the overall characteristics and the thin vessel features preserving the capillaries from fundus images. Then, considering that the single-channel network is unsuitable for the unified characterization of thick and thin vessels, we develop a dual-channel asymmetric CNN based on the U-Net model. The MainSegment-Net uses the step-by-step connection mode to achieve rapid positioning and segmentation of thick vessels; the FineSegment-Net combines dilated convolution and the skip connection to achieve the fine extraction of thin vessels. Finally, the output of the dual-channel asymmetric CNN is fused and coded to combine the segmentation results of thick and thin vessels. The performance of our method is evaluated and tested by DRIVE and CHASE_DB1. The results show that the accuracy (Acc), sensitivity (SE), and specificity (SP) of our method on the DRIVE database are 0.9630, 0.8745, and 0.9823, respectively. The evaluation indexes Acc, SE, and SP of the CHASE_DB1 database are 0.9694, 0.8916, and 0.9794, respectively. Additionally, our method combines the biological vision mechanism with deep learning to achieve rapid and automatic segmentation of retinal vessels, providing a new idea for diagnosing and analyzing subsequent medical images.

2

A dark and bright channel prior guided deep network for retinal image quality assessment

Xu Ziwen, Zou Beiji, Liu Qing

Biocybernetics and Biomedical Engineering

|

2022

|

Vol. 42, no. 3

772--783

EN

Background and objective: Retinal image quality assessment is an essential task for the diagnosis of retinal diseases. Recently, there are emerging deep models to grade quality of retinal images. However, current models either directly transfer classification networks originally designed for natural images to quality classification of retinal images or introduce extra image quality priors via multiple CNN branches or independent CNNs. The purpose of this work is to address retinal image quality assessment by a simple deep model. Methods: We propose a dark and bright channel prior guided deep network for retinal image quality assessment named GuidedNet. It introduces dark and bright channel priors into deep network without extra parameters increasing and allows for training end-to-end. In detail, the dark and bright channel priors are embedded into the start layer of a deep network to improve the discriminate ability of deep features. Moreover, we re-annotate a new retinal image quality dataset called RIQA-RFMiD for further validation. Results: The proposed method is evaluated on a public retinal image quality dataset Eye-Quality and our re-annotated dataset RIQA-RFMiD. We obtain the average F-score of 88.03% on Eye-Quality and 66.13% on RIQA-RFMiD, respectively. Conclusions: We investigate the utility of the dark and bright channel priors for retinal image quality assessment. And we propose a GuidedNet by embedding the dark and bright channel priors into CNNs without much model burden. Moreover, to valid the GuidedNet, we re-create a new dataset RIQA-RFMiD. With the GuidedNet, we achieves state-of-the-art performances on a public dataset Eye-Quality and our re-annotated dataset RIQA-RFMiD.

3

A hybrid method for blood vessel segmentation in images

Díaz Primitivo, Rodríguez Alma, Cuevas Erik, Valdivia Arturo, Chavolla Edgar, Pérez-Cisneros Marco, Zaldívar Daniel

Biocybernetics and Biomedical Engineering

|

2019

|

Vol. 39, no. 3

814--824

EN

In the last years, image processing has been an important tool for health care. The analysis of retinal vessel images has become crucial to achieving a better diagnosis and treatment for several cardiovascular and ophthalmological deceases. Therefore, an automatic and accurate procedure for retinal vessel and optic disc segmentation is essential for illness detection. This task is extremely hard and time-consuming, often requiring the assistance of human experts with a high degree of professional skills. Several retinal vessel segmentation methods have been developed with satisfactory results. Nevertheless, most of such techniques present a poor performance mainly due to the complex structure of vessels in retinal images. In this paper, an accurate methodology for retinal vessel and optic disc segmentation is presented. The proposed scheme combines two different techniques: the Lateral Inhibition (LI) and the Differential Evolution (DE). The LI scheme produces a new image with enhanced contrast between the background and retinal vessels. Then, the DE algorithm is used to obtain the appropriate threshold values through the minimization of the cross-entropy function from the enhanced image. To evaluate the performance of the proposed approach, several experiments over images extracted from STARE, DRIVE, and DRISHTI-GS databases have been conducted. Simulation results demonstrate a high performance of the proposed scheme in comparison with similar methods reported in the literature.

4

Fast, accurate and robust retinal vessel segmentation system

Jiang Z., Yepez J., An S., Ko S.

Biocybernetics and Biomedical Engineering

|

2017

|

Vol. 37, no. 3

412--421

EN

The accurate segmentation of the retinal vessel tree has become the prerequisite step for automatic ophthalmological and cardiovascular diagnosis systems. Aside from accuracy, robustness and processing speed are also considered crucial for medical purposes. In order to meet those requirements, this work presents a novel approach to extract blood vessels from the retinal fundus, by using morphology-based global thresholding to draw the retinal venule structure and centerline detection method for capillaries. The proposed system is tested on DRIVE and STARE databases and has an average accuracy of 95.88% for single-database test and 95.27% for the cross-database test. Meanwhile, the system is designed to minimize the computing complexity and processes multiple independent procedures in parallel, thus having an execution time of 1.677 s per image on CPU platform.

5

Automatic parameters selection of Gabor filters with the imperialism competitive algorithm with application to retinal vessel segmentation

Farokhian F., Yang C., Demirel H., Wu S., Beheshti I.

Biocybernetics and Biomedical Engineering

|

2017

|

Vol. 37, no. 1

246--254

EN

Retinal images play an important role in the early diagnosis of diseases such as diabetes. In the present study, an automatic image processing technique is proposed to segment retinal blood vessels in fundus images. The technique includes the design of a bank of 180 Gabor filters with varying scale and elongation parameters. Furthermore, an optimization method, namely, the imperialism competitive algorithm (ICA), is adopted for automatic parameter selection of the Gabor filter. In addition, a systematic method is proposed to determine the threshold value for reliable performance. Finally, the performance of the proposed approach is analyzed and compared with that of other approaches on the basis of the publicly available DRIVE database. The proposed method achieves an area under the receiver operating characteristic curve of 0.953 and an average accuracy of up to 0.9392. Thus, the results show that the proposed method is well comparable with alternative methods in the literature.

6

A retinal image authentication framework based on a graph-based representation algorithm in a two-stage matching structure

Khakzar M., Pourghassem H.

Biocybernetics and Biomedical Engineering

|

2017

|

Vol. 37, no. 4

742--759

EN

Retinal vascular pattern has many valuable characteristics such as uniqueness, stability and permanence as a basis for human authentication in security applications. This paper presents an automatic rotation-invariant retinal authentication framework based on a novel graph-based retinal representation scheme. In the proposed framework, to replace the retinal image with a relational mathematical graph (RMG), we propose a novel RMG definition algorithm from the corresponding blood vessel pattern of the retinal image. Then, the unique features of RMG are extracted to supplement the authentication process in our framework. The authentication process is carried out in a two-stage matching structure. In the first stage of this scenario, the defined RMG of enquiry image is authenticated with enrolled RMGs in the database based on isomorphism theory. If the defined RMG of enquiry image is not isomorphic with none enrolled RMG in the database, in the second stage of our matching structure, the authentication is performed based on the extracted features from the defined RMG by a similarity-based matching scheme. The proposed graph-based authentication framework is evaluated on VARIA database and accuracy rate of 97.14% with false accept ratio of zero and false reject ratio of 2.85% are obtained. The experimental results show that the proposed authentication framework provides the rotation invariant, multi resolution and optimized features with low computational complexity for the retina-based authentication application.

7

Box-counting fractal dimension in application to recognition of hypertension through the retinal image analysis

Kruk M., Świderski B., Osowski S.

Przegląd Elektrotechniczny

|

2013

|

R. 89, nr 9

286-289

EN

The paper presents the application of the box-counting dimension to the recognition of hypertension through the analysis of the image of the eye fundus. The box-counting dimension represents a single measure, often used to describe the structure of fractal-like images. We propose based on it fast method of classification of the retinal image in order to recognize the class of healthy, introductory step and advanced illness cases. The results of experiments performed on 125 cases confirm good performance of the proposed method.

PL

Praca prezentuje zastosowanie wymiaru pudełkowego do rozpoznania zmian ciśnieniowych poprzez automatyczna analizę obrazu dna oka. Zaproponowana została metoda szybkiego rozpoznania stanu chorobowego rozróżniająca trzy klasy: zdrowi, początkowy stan choroby i stan zaawansowany. Przedstawione są rezultaty rozpoznania dotyczące 125 przypadków.

8

An automatic hybrid method for retinal blood vessel extraction

Yang Y., Huang S., Rao N.

International Journal of Applied Mathematics and Computer Science

|

2008

|

Vol. 18, no 3

399-407

EN

The extraction of blood vessels from retinal images is an important and challenging task in medical analysis and diagnosis. This paper presents a novel hybrid automatic approach for the extraction of retinal image vessels. The method consists in the application of mathematical morphology and a fuzzy clustering algorithm followed by a purification procedure. In mathematical morphology, the retinal image is smoothed and strengthened so that the blood vessels are enhanced and the background information is suppressed. The fuzzy clustering algorithm is then employed to the previous enhanced image for segmentation. After the fuzzy segmentation, a purification procedure is used to reduce the weak edges and noise, and the final results of the blood vessels are consequently achieved. The performance of the proposed method is compared with some existing segmentation methods and hand-labeled segmentations. The approach has been tested on a series of retinal images, and experimental results show that our technique is promising and effective.