Digital Image Processing
A Novel Algorithm for Extraction of the Layers of the Cornea
CRV '09 Proceedings of the 2009 Canadian Conference on Computer and Robot Vision
| Hi-index | 0.00 |
This paper presents a novel processing technique which can be applied to corneal in vivo images obtained with optical coherence tomograms across the central meridian of the cornea. The method allows to estimate the thickness of the corneal sublayers (Epithelium, Bowman's layer, Stroma, Endothelium, and whole corneal thickness) at any location, including the center and the midperiphery, on both nasal and temporal sides. The analysis is carried out on both the pixel and subpixel scales to reduce the uncertainty in thickness estimations. This technique allows quick and noninvasive assessment of patients. As an example of application and validation, we present the results obtained from the analysis of 52 healthy subjects, each with 3 scans per eye, for a total of more than 300 images. Particular attention has been paid to the statistical interpretation of the obtained results to find a representative assessment of each sublayer's thickness.