No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Segmentation of retinal vessels in adaptive optics images for assessment of vasculitis
... Some sort of consecutive convolution cellular levels and also pooling cellular levels keep to the input data, so your multilevel can certainly find out the fuzy features so that you can segment retinal vessel. Rossant et al [11] has presented the approach which is committed by using substantial improvements resulting from vasculitis, however it is as well genuine for vessel segmentation by using average need of alteration. That uses a presegmentation measure which is essential for any robustness in addition to accuracy and reliability of your results. ...
Adaptive optics (AO) fundus imaging is an optoelectronic technique allowing an improvement of an order of magnitude of lateral resolution of retinal images. Currently, its main applications in ophthalmology span from photoreceptor to retinal pigment epithelial cells and vessels, each of them being affected by specific diseases. Technological and image processing improvements are expanding the scope of its medical applications. Here we will review some of the current and envisioned applications of AO in clinical practice.
In this paper, we propose two important improvements of an existing approach for automatically segmenting the walls of retinal arter-ies of healthy / pathological subjects in adaptive optics images. We illus-trate the limits of the previous approach and propose to (i) modify the pre-segmentation step, and (ii) embed additional information through coupling energy terms in the parallel active contour model. The interest of these new elements as well as the pre-segmentation step is then evalu-ated against manual segmentations. They improve the robustness against low contrasted walls and morphological deformations that occur along vessels in case of pathologies. Noticeably, this strategy permits to obtain a mean error of 13.4% compared to an inter-physicians error of 17%, for the wall thickness which is the most sensitive measure used. Additionally, this mean error is in the same range than for healthy subjects.
Parametric deformable models are an important technique for image segmentation. In order to improve the robustness of the model, it may be interesting to incorporate a priori information about the shape of the objects to be segmented. In this paper, we propose to add a parallelism constraint. Such a model is relevant in many applications where elongated structures have to be detected. One main advantage of our formulation is that it only needs few parameters to be adjusted in addition to those of traditional snakes. The proposed model has been applied for the segmentation of OCT images of the retina and for the segmentation of retinal vessels. Experimental results, obtained on 25 OCT images and 30 eye fundus images, demonstrated the robustness, flexibility and large potential applicability of this new formulation. The accuracy of the method has been assessed by comparing manual segmentations, made by experts, with the automatic ones.
Even when corrected with the best spectacles or contact lenses, normal human eyes still suffer from monochromatic aberrations that blur vision when the pupil is large. We have successfully corrected these aberrations using adaptive optics, providing normal eyes with supernormal optical quality. Contrast sensitivity to fine spatial patterns was increased when observers viewed stimuli through adaptive optics. The eye's aberrations also limit the resolution of images of the retina, a limit that has existed since the invention of the ophthalmoscope. We have constructed a fundus camera equipped with adaptive optics that provides unprecedented resolution, allowing the imaging of microscopic structures the size of single cells in the living human retina.
Snakes, or active contours, are used extensively in computer
vision and image processing applications, particularly to locate object
boundaries. Problems associated with initialization and poor convergence
to boundary concavities, however, have limited their utility. This paper
presents a new external force for active contours, largely solving both
problems. This external force, which we call gradient vector flow (GVF),
is computed as a diffusion of the gradient vectors of a gray-level or
binary edge map derived from the image. It differs fundamentally from
traditional snake external forces in that it cannot be written as the
negative gradient of a potential function, and the corresponding snake
is formulated directly from a force balance condition rather than a
variational formulation. Using several two-dimensional (2-D) examples
and one three-dimensional (3-D) example, we show that GVF has a large
capture range and is able to move snakes into boundary concavities
Adaptive optics imaging of the retina has recently proven its capability to image micrometric structures such as blood vessels, involved in common ocular diseases. In this paper, we propose an approach for automatically segmenting the walls of retinal arteries in the images acquired with this technology. The walls are modeled as four curves approximately parallel to a previously detected reference line located near the vessel center (axial reflection). These curves are first initialized using a tracking procedure and then more accurately positioned using an active contour model embedding a parallelism constraint. We consider both healthy and pathological subjects in the same framework and show that the proposed method applies in all cases. Extensive experiments are also proposed, by analyzing the robustness of the axial reflections detection, the influence of the tracking parameters as well as the performance of the tracking and the active contour model. Noticeably, the results show a good robustness for detecting axial reflections and a moderate influence of the tracking parameters. Compared to a naive initialization, the active contour model coupled with the tracking also offers faster convergence and better accuracy while keeping an overall error smaller or very near the inter-physicians error.
Purpose To document perivascular infiltrates in the retina using high resolution flood illumination adaptive optics (AO) imaging in patients with retinal vasculitis.
Methods The charts of five subjects with retinal vasculitis showing evidence of perivascular infiltrates by means of AO imaging (rtx1 camera, Imagine Eye, Orsay, France) were reviewed.
Results Diagnoses included primary idiopathic retinal vasculitis (n=1), Lyme’s disease (n=2), TB related retinal vasculitis (n=1) and idiopathic retinitis, vasculitis and aneurysm (IRVAN; n=1). Using AO imaging, perivascular infiltrates appeared as discrete, linear areas of opacification along the veins, often surrounding the venous narrowing. AO imaging was more sensitive that either fundus photographs or fluorescein angiography for their detection. Follow‐up examinations (n=3) showed a changing pattern of narrowing of the vessel lumen, and eventual disappearance of these perivascular infiltrates. In the IRVAN case, inflammatory infiltrates were found around arterioles and veins as well, suggesting an inflammation induced‐phenomenon.
Conclusion AO imaging has a higher sensitivity than fundus photographs or fluorescein angiography to detect perivascular inflammatory infiltrates. Vascular inflammation is strongly linked to changes in vascular lumen diameter. AO imaging may have a particular interest when diagnosis and following subjects with retinal vasculitis.
Retinal images sequences provided by adaptive optics instruments have to be processed before clinical exploitation. We present a robust procedure that accounts for non-translational motions and variable image quality to deliver improved reconstructed images.
This research is aimed at characterizing in vivo differences between healthy and pathological retinal tissues at the microscopic scale using a compact adaptive optics (AO) retinal camera. Tests were performed in 120 healthy eyes and 180 eyes suffering from 19 different pathological conditions, including age-related maculopathy (ARM), glaucoma and rare diseases such as inherited retinal dystrophies. Each patient was first examined using SD-OCT and infrared SLO. Retinal areas of 4°x4° were imaged using an AO flood-illumination retinal camera based on a large-stroke deformable mirror. Contrast was finally enhanced by registering and averaging rough images using classical algorithms. Cellular-resolution images could be obtained in most cases. In ARM, AO images revealed granular contents in drusen, which were invisible in SLO or OCT images, and allowed the observation of the cone mosaic between drusen. In glaucoma cases, visual field was correlated to changes in cone visibility. In inherited retinal dystrophies, AO helped to evaluate cone loss across the retina. Other microstructures, slightly larger in size than cones, were also visible in several retinas. AO provided potentially useful diagnostic and prognostic information in various diseases. In addition to cones, other microscopic structures revealed by AO images may also be of interest in monitoring retinal diseases.
Alterations in leukocyte velocity have been implicated in many retinal disease processes. However, direct and objective assessment of leukocyte velocity in retinal capillaries has been limited by a reliance on invasive contrast dyes that allow leukocyte visualization only for a short time span. The recent application of adaptive optics in a scanning laser ophthalmoscope (AOSLO) has made long-term imaging of parafoveal leukocyte movement possible without contrast dyes. In this study, using the AOSLO, we demonstrate a new method to investigate retinal parafoveal capillary leukocyte velocity.
Experimental study.
Six normal healthy subjects ranging from 25 to 35 years of age with clear ocular media.
The parafoveal zone of the retina was imaged in all subjects using an AOSLO.
Leukocyte velocity was determined in the parafoveal capillaries including the foveal avascular zone border. Leukocyte velocity was measured directly from movie segments in which the leukocytes were clearly visible.
The mean parafoveal leukocyte velocity for 6 subjects was 1.37 mm/second, ranging from 0.77 to 2.10 mm/second. Leukocytes were not visible in all parafoveal capillaries.
Parafoveal capillary leukocyte velocity can be directly and noninvasively measured without the use of contrast dyes using an AOSLO.
Segmentation of Retinal Arteries in Adaptive Optics Images
- N Lermé
- F Rossant
- I Bloch
- M Paques
- E Koch
Snakes, shapes and gradient vector flow
- C Xu
- J Prince
Adaptive Optics imaging patterns of retinal vasculitis
- M.-H Errera
- J Benesty
- E Koch
- C Chaumette
- J A Sahel
- B Bodaghi
- M Paques