Caserel

Caserel segments retinal layers in Spectral Domain optical coherence tomography (OCT) images to delineate anatomical layer boundaries for ophthalmic analysis and diagnostic support.

Key Features:

• Automated segmentation: Automates segmentation of retinal layers in Spectral Domain OCT images.
• Graph theory: Applies graph theory to model retinal layer boundaries.
• Dynamic programming: Uses dynamic programming to identify optimal boundary paths.
• Eight boundary delineation: Delineates eight distinct retinal layer boundaries in normal adult eyes.
• Validation against expert graders: Segmentation results were compared with expert graders and showed closer alignment to one expert grader than to a second grader.
• Implementation: Implemented using Matlab.

Scientific Applications:

• Ophthalmic diagnosis: Supports diagnostic accuracy and efficiency in ophthalmology by providing consistent retinal layer segmentation.
• Ocular disease study: Facilitates analysis of anatomical and pathological structures relevant to various ocular diseases.
• Research and clinical imaging: Enables quantitative analysis in research and clinical applications of ophthalmic imaging using Spectral Domain OCT.

Methodology:

Caserel employs graph theory and dynamic programming to automate segmentation of retinal layer boundaries in Spectral Domain OCT images, delineating eight boundaries in normal adult eyes; the algorithms were implemented in Matlab and validated by comparative analysis with expert graders.