Segmentation of Retinal Arteries in Adaptive Optics Images

Abstract

Abstract-In this paper, we present a method for automatically segmenting the walls of retinal arteries in adaptive optics images. To the best of our knowledge, this is the first method addressing this problem in such images. To achieve this goal, we propose to model these walls as four curves approximately parallel to a common reference line located near the center of vessels. Once this line is detected, the curves are simultaneously positioned as close as possible to the borders of walls using an original tracking procedure to cope with deformations along vessels. Then, their positioning is refined using a deformable model embedding a parallelism constraint. Such an approach enables us to control the distance of the curves to their reference line and improve the robustness to image noise. This model was evaluated on healthy subjects by comparing the results against segmentations from physicians. Noticeably, the error introduced by this model is smaller or very near the inter-physicians error.

Venous Nicking Without Arteriovenous Contact

“…Supplemental content at jamaophthalmology.com Arteries and veins were semiautomatically segmented from AO images using a custom software running under Matlab (MathWorks). 10 The segmentation algorithm is based on the detection of 4 curves parallel to the central reflection, which are refined using a deformable model incorporating a parallelism constraint. Graphic representations of the venous diameter and of the width of the intervascular space (ie, the distance between the venous lumen and the outer limit of the arteriolar wall) along a given venous segment were generated.…”

Adaptive optics imaging of retinal microstructures: Image processing for medical applications

“…energy functional, that defines the active makes evolve independently four curves to wall contours assuming that they are almo axial reflection. In [11], other structural fe modelled and integrated, in order to improv against low contrasted walls and deformations that occur along vessels in cas For this, new coupled energy terms wer energy functional, modeling symetry prop side of the axial reflection. In this new curves do no evolve independently any m The strength of parallelism and symetr controlled through weighting parameters optimized experimentally so as to cover th of images and pathologies.…”

“…rteriole (between ood illumination proposed by our semi-automatical oles [10,11]. Our i) detection and flection, (ii) prere relying mostly elineation of the contour model .…”

Segmentation of tumor vessels based on parallel double snakes including region information

“…in many imaging modalities, see for example the review in [13]. They exploit the panel of tools available in the image processing field and adapt them to the application specificity: vessel enhancement based on wavelets [21], morphological filter [24], Hessianbased filters [1], [22], [14], pixel classification [22], [21], region growing algorithms [9], [14], [16], tracking procedures [8], active contour models [15], [12], etc. However, to the best of our knowledge, very little work has been dedicated to the analysis of tumor vessels.…”

“…The ISEP team has developed recently a new model, the Parallel Double Snakes, that incorporates a parallelism constraint in the energy function, allowing the simultaneous extraction of two almost parallel contours [20]. It has been succesfully applied for the segmentation of vessels in adaptive optics [12]. This approach can be coupled with a centerline extraction method [4], [5], in order to initialize the snakes.…”