[Automated segmentation of retina layer structures on optical coherence tomography]

Area of Science:

• Ophthalmology
• Biomedical Imaging
• Image Processing

Context:

• Accurate segmentation of retinal layers in optical coherence tomography (OCT) is crucial for diagnosing diseases like glaucoma and retinopathy.
• Current OCT image analysis methods often lack reliability and require high-quality images, limiting their clinical utility.
• Developing robust algorithms for automated retinal layer segmentation is essential for advancing early disease detection.

Purpose:

• To develop and validate an improved algorithm for automatic segmentation of retinal layers using optical coherence tomography (OCT) images.
• To address limitations of existing methods by employing advanced image processing techniques, including improved complex nonlinear diffusion filtering.
• To accurately detect boundaries of key retinal layers and measure photoreceptor layer thickness for quantitative analysis.

Summary:

• The study utilized an algorithm incorporating automatic thresholding, improved complex nonlinear diffusion filtering, morphological operations, and peak detection for retinal layer segmentation.
• The algorithm was applied to 20 retinal OCT images, accurately segmenting the inner limiting membrane (ILM), outer nuclear layer (ONL), photoreceptor segments (IS/OS), and RPE_ChCap layer.
• Photoreceptor layer thickness was measured, demonstrating strong correlation with expert manual segmentation and consistency with established OCT measurements.

Impact:

• The developed algorithm shows promise for prospective application in the clinical diagnosis of retinal diseases.
• Accurate and reliable automated segmentation of retinal layers can significantly improve the efficiency and accuracy of diagnosing conditions like glaucoma and retinopathy.
• This advancement in OCT image analysis contributes to the early detection and management of sight-threatening retinal disorders.