Graph-based segmentation of retinal layers in OCT images

Graph-based segmentation of retinal layers in OCT images segments retinal layers from optical coherence tomography (OCT) images to delineate seven retinal boundaries and enable detailed structural analysis in ophthalmology.

Key Features:

• Graph-based algorithm: Uses graph-based algorithms to delineate multiple retinal boundaries in OCT images.
• Implementation: Developed in Matlab.
• Segmentation capabilities: Automatically segments six retinal layers by identifying seven key boundaries: Inner Limiting Membrane (ILM); Nerve Fiber Layer/Ganglion Cell Layer (NFL/GCL); Inner Plexiform Layer/Inner Nuclear Layer (IPL/INL); Inner Nuclear Layer/Optical Plexiform Layer (INL/OPL); Outer Plexiform Layer/Outer Nuclear Layer (OPL/ONL); Isocortex/Synaptic Layer (IS/OS); and Retinal Pigment Epithelium (RPE).

Scientific Applications:

• Retinal imaging analysis: Provides precise delineation of retinal layer boundaries in OCT images for structural assessment.
• Ophthalmology research and clinical assessment: Enables quantitative analysis of retinal layers for research and clinical studies in ophthalmology.

Methodology:

Implemented in Matlab and employing graph-based segmentation algorithms to automatically identify seven retinal layer boundaries corresponding to six retinal layers.