Cellular Neural Networks With Virtual Template Expansion for Retinal Vessel Segmentation

Perfetti, R.; Ricci, Elisa; Casali, Daniele; Costantini, Giovanni

doi:10.1109/tcsii.2006.886244

Cited by 71 publications

(37 citation statements)

References 16 publications

(32 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The results with TABLE III RESULTS ON THE COMPLETE STARE DATABASE TABLE IV RESULTS ON THE COMPLETE DRIVE DATABASE the transform (denoted as Proposed Method) and without the transform, which is in the original image space (denoted as Proposed Method (ImS)), are shown in the tables. The comparative results of the benchmark methods in the tables are obtained from [34]- [35], [37]- [41]. For the DRIVE database, there are two different results for the line detector method as reported in [34] and [35].…”

Section: Resultsmentioning

confidence: 99%

“…The comparative results of the benchmark methods in the tables are obtained from [34]- [35], [37]- [41]. For the DRIVE database, there are two different results for the line detector method as reported in [34] and [35]. Both of them are shown in Table IV for reference together with the result from our own implementation of the line detector algorithm of [35], denoted as Line (Impl.).…”

Section: Resultsmentioning

confidence: 99%

“…For the DRIVE database, there are two different results for the line detector method as reported in [34] and [35]. Both of them are shown in Table IV for reference together with the result from our own implementation of the line detector algorithm of [35], denoted as Line (Impl.). Ideally, the results of Line (Impl.)…”

Section: Resultsmentioning

confidence: 99%

“…This kind of line shapes is shown to be robust for detecting linear structure [34], [35], [48]. is the space of square shapes.…”

Section: Line-shape Concavity and Locally Normalized Concavitymentioning

confidence: 99%

“…In each level, the vessels are extracted using the intensity, angle and length information based on a predefined elongated concave structure. Ricci and Perfetti [34], [35] proposed a fast and robust template matching algorithm which measures the output difference between a line filter and an averaging filter. Tolis et al [36] proposed a tracking approach in which seed points are grown according to predefined rules to extract all concave regions as vessels.…”

mentioning

confidence: 99%

See 4 more Smart Citations

General Retinal Vessel Segmentation Using Regularization-Based Multiconcavity Modeling

Lam

Gao

Liew

2010

IEEE Trans. Med. Imaging

241

105

View full text Add to dashboard Cite

Abstract-Detecting blood vessels in retinal images with the presence of bright and dark lesions is a challenging unsolved problem. In this paper, a novel multiconcavity modeling approach is proposed to handle both healthy and unhealthy retinas simultaneously. The differentiable concavity measure is proposed to handle bright lesions in a perceptive space. The line-shape concavity measure is proposed to remove dark lesions which have an intensity structure different from the line-shaped vessels in a retina. The locally normalized concavity measure is designed to deal with unevenly distributed noise due to the spherical intensity variation in a retinal image. These concavity measures are combined together according to their statistical distributions to detect vessels in general retinal images. Very encouraging experimental results demonstrate that the proposed method consistently yields the best performance over existing state-of-the-art methods on the abnormal retinas and its accuracy outperforms the human observer, which has not been achieved by any of the state-of-the-art benchmark methods. Most importantly, unlike existing methods, the proposed method shows very attractive performances not only on healthy retinas but also on a mixture of healthy and pathological retinas.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

“…This kind of line shapes is shown to be robust for detecting linear structure [34], [35], [48]. is the space of square shapes.…”

Section: Line-shape Concavity and Locally Normalized Concavitymentioning

confidence: 99%

mentioning

confidence: 99%

See 3 more Smart Citations

General Retinal Vessel Segmentation Using Regularization-Based Multiconcavity Modeling

Lam

Gao

Liew

2010

IEEE Trans. Med. Imaging

241

105

View full text Add to dashboard Cite

show abstract

Split and shift methodology on cellular processor arrays: area saving versus time penalty

Fernandez

Brea

Cabello

2012

Circuit Theory & Apps

View full text Add to dashboard Cite

This paper addresses the so-called split and shift methodology. This methodology deals with the implementation of kernels of sizes that go above the physically implemented resources (local connections and weighting circuits) on synchronous cellular processor arrays (CPA), including the realization of large neighborhood operations and/or the reduction of the available hardware in order to drop the area consumption. Two main goals are pursued in the development of the methodology, namely: (1) minimum penalty at processing time and (2) absolutely no penalty at functional level. The paper presents different techniques and guidelines for the methodology application and introduces a Figure of Merit to evaluate them by relating area gains with time penalty. This, along with a kernel shape analysis, led us to propose more adequate configurations of weighting circuits and to justify the classical choice of North-East-West-South connectivity. To validate the methodology, we realize several estimates over actual physical implementations, and we propose the realization over CPAs of the spin filters, scale invariant feature transform and speeded-up robust features algorithms. A more in-depth trade-off analysis is realized over the implementation of the pixel level snakes algorithm.but it can interact with separated PEs, thanks to the propagative effects of the array dynamics of kernel application. The basic characteristic of local connectivity and its simple SIMD control makes this kind of systems very suitable for hardware implementation. However, as a consequence, the natural size of the kernels to be applied is limited to the smallest one (3 Â 3). This is, on the other hand, an important limitation in the functionality of a CPA as larger neighborhoods are needed in several image processing primitives as diffusion or low-pass filtering operations [1], halftoning [2], texture analysis [3] or matching and hit&miss operations [4], some of them used in algorithms like modern scale-and rotation-invariant feature extractors like scale invariant feature transform (SIFT) and speeded-up robust features (SURF) [5,6].As indicated, a CPA can realize global processing taking into account the whole image information thanks to the propagative effects of the architecture. According to this we can think in solving the remote neighbors interaction through the recursive application of templates. In fact, a recursive process can be summarized in the application of a large neighborhood (LN) template. Nevertheless, the inverse process, the decomposition of a LN template into minimum-sized templates is not trivial, and different approaches have been developed to deal with this issue.The challenge is then to implement any kind of LN operations while keeping the local connectivity and with affordable penalties in performance. Our goal is to do it, in addition, with the minimum impact in the architecture at hardware level.On the other hand, on focal-plane processors with a pixel-to-PE assignment, the area occupation is not only a matter of cost or h...

show abstract

Retinal blood vessel segmentation for macula detachment surgery monitoring instruments

Hajabdollahi

Karimi

Soroushmehr

et al. 2018

Circuit Theory & Apps

View full text Add to dashboard Cite

In recent years image processing has improved detection and diagnosis in medical application. Image processing applications are now embedded in medical instruments such as MRI and CT. In the case of retinopathy, fast extraction of blood vessels can allow the physician to view injury regions during surgery. Macula detachment surgeries, or computer-aided intraocular surgeries, require precise and real-time knowledge of the vasculature during the operation. Use of artificial neural network has produced good results in image processing applications, but its implementation may not be suitable for real-time applications in small, embedded hardware. Because of error resiliency of the neural network, its structure can be pruned and simplified. In this paper an efficient hardware implementation of neural network for retinal vessel segmentation is proposed. We simplify the neural network structure in such a way that the accuracy of the results is not altered significantly. Simulation results and FPGA implementation show that our proposed network has low complexity and can be applied for segmentation of retinal vessels with acceptable accuracy. This makes the proposed method a good candidate to be implemented in any device such as a binocular indirect ophthalmoscope. KEYWORDS artificial neural networks (ANN), error compensation, hardware implementation, retinal vessel segmentation, retinopathy

show abstract

Cellular Neural Networks With Virtual Template Expansion for Retinal Vessel Segmentation

Cited by 71 publications

References 16 publications

General Retinal Vessel Segmentation Using Regularization-Based Multiconcavity Modeling

General Retinal Vessel Segmentation Using Regularization-Based Multiconcavity Modeling

Split and shift methodology on cellular processor arrays: area saving versus time penalty

Retinal blood vessel segmentation for macula detachment surgery monitoring instruments

Contact Info

Product

Resources

About