期刊名称：International Journal of Computer Science and Network
印刷版ISSN：2277-5420
出版年度：2012
卷号：1
期号：3
出版社：IJCSN publisher
摘要：Segmentation of blood vessels in retinal images providesearly diagnosis of diseases like glaucoma, diabeticretinopathy and macular degeneration. Among thesediseases occurrence of Glaucoma is most frequent and hasserious ocular consequences that can even lead toblindness, if it is not detected early. The clinical criteria forthe diagnosis of glaucoma include intraocular pressuremeasurement, optic nerve head evaluation, retinal nervefiber layer and visual field defects. This form of bloodvessel segmentation helps in early detection for ophthalmicdiseases, and potentially reduces the risk of blindness.The low-contrast images at the retina owing to narrowblood vessels of the retina are difficult to extract. Theselow contrast images are, however useful in revealingcertain systemic diseases. Motivated by the goals ofimproving detection of such vessels, this present workproposes an algorithm for segmentation of blood vessels,and compares the results between expert ophthalmologists’hand-drawn ground-truths and segmented image (i.e. theoutput of the present work). Sensitivity, specificity, positivepredictive value (PPV), positive likelihood ratio (PLR) andaccuracy are used to evaluate overall performance. It isfound that this work segments blood vessels successfullywith sensitivity, specificity, PPV, PLR and accuracy of99.62%, 54.66%, 95.08%, 219.72 and 95.03%,respectively.
