Detection and Segmentation of Retinal Lesions in Retcam 3 Images Based on Active Contours Driven by Statistical Local Features

Recent Techniques and Trends for Retinal Blood Vessel Extraction and Tortuosity Evaluation: A Comprehensive Review

“…These are images of premature infants, so, among other things, there is a large number of choroidal vessels or artefacts caused by the movement of the child's eyes. The dataset clearly shows the gestational age of the child at the time of birth and their birth weight [11,36,37].…”

“…These procedures have the potential to optimise the effectivity and robustness of segmentation procedures. The first step in pre-processing is to convert the colour image to greyscale, the next steps are: suppression of image noise (image filtering), improving contrast and brightness transformation [11,36,39,40,41].…”

“…The fuzzy edge detector uses the method of Kubicek et al [36]. for the extraction of retinal blood vessels, which uses a combination of fuzzy logic and morphological techniques.…”

“…The fuzzy edge detector detects edges and suppresses high-frequency noise in a non-contrast image. A binary model of retinal blood vessels is subsequently obtained by morphological operations [36].…”

“…Tortuosity, or vessel curvature, is a type of vascular pathology. This is one of the important parameters for determining the presence and severity of various diseases [36]. It affects both arteries and veins, with slight curvature of blood vessels without clinical symptoms commonly observed in both humans and animals [21].…”