Automated detection of the retinal from OCT spectral domain images of healthy eyes

Gaspare Giovinco, Maria Cristina Savastano, Salvatore Ventre, Antonello Tamburrino

Risultato della ricerca: Contributo in rivistaArticolo in rivista


Optical coherence tomography (OCT) has become one of the most relevant diagnostic tools for retinal diseases. Besides being a non-invasive technique, one distinguished feature is its unique capability of providing (in vivo) cross-sectional view of the retinal. Specifically, OCT images show the retinal layers. From the clinical point of view, the identification of the retinal layers opens new perspectives to study the correlation between morphological and functional aspects of the retinal tissue. The main contribution of this paper is a new method/algorithm for the automated segmentation of cross-sectional images of the retina of healthy eyes, obtained by means of spectral domain optical coherence tomography (SD-OCT). Specifically, the proposed segmentation algorithm provides the automated detection of different retinal layers. Tests on experimental SD-OCT scans performed by three different instruments/manufacturers have been successfully carried out and compared to a manual segmentation made by an independent ophthalmologist, showing the generality and the effectiveness of the proposed method.
Lingua originaleEnglish
pagine (da-a)837-850
Numero di pagine14
RivistaJournal of Modern Optics
Stato di pubblicazionePubblicato - 2015


  • biomedical image processing
  • image segmentation
  • macula
  • ophthalmology
  • optical coherence tomography
  • retinal layers


Entra nei temi di ricerca di 'Automated detection of the retinal from OCT spectral domain images of healthy eyes'. Insieme formano una fingerprint unica.

Cita questo