Comparison of Edge Detection Techniques on Vessel Detection of Infant's Retinal Image

• Autores: Lassada Sukkaew, Bunyarit Uyyanonvara, Sarah Boman
• Localización: International journal of the computer, the internet and management, ISSN 0858-7027, Vol. 13, Nº. 2 (OCT), 2005 (Ejemplar dedicado a: Suplemento 2: eIndustry 2005. Proceedings of the International Conference on Computer and Industrial Management), págs. 6-6
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • ROP or Retinopathy of Prematurity is abnormal growth of blood vessels in the baby's eye. It is a sight threatening condition where the process of retinal vessel growth is disrupted by preterm birth and some of its complications. The earlier the symptom can be detected, the easier the patient can be treated. One of identifiers of an early stage of symptom detection is retinal blood vessel. Even though there are many works describing successful applications of image analysis techniques to adult retinal images, most of the techniques failed when applied on images of obscured and not-welldeveloped infant's retinal blood vessel. This article proposes result of a comparison of edge detection techniques for automatic classification of retinal blood vessels in infant images. The accuracy of methods are examined by comparing the results with hand-drawn ground-truth images. The finding indicats that the sensitivity and specificity of highest results in the first group of operators is Canny operator are 50.08% and 64.77%. Laplacian of Gaussian show best result of 64.81% and 64.23% for a set of 20 images.