New variational image decomposition model for simultaneously denoising and segmenting optical coherence tomography...

Area of Science:

• Medical Imaging
• Image Processing
• Ophthalmology

Background:

• Optical coherence tomography (OCT) is crucial for retinal disease diagnosis and monitoring.
• Automated OCT image analysis is challenging due to inherent image noise.
• Accurate segmentation of retinal layers is essential for clinical applications.

Purpose of the Study:

• To propose a novel variational image decomposition model for OCT images.
• To effectively remove noise and extract retinal layer boundaries from OCT images.
• To enhance the computational efficiency of the proposed decomposition model.

Main Methods:

• A variational image decomposition model is introduced to separate OCT images into denoised, edge, and texture components.
• A fast Fourier transform-based split Bregman algorithm is developed for efficient model solving.
• The model is validated using both synthesized and real-world OCT datasets.

Main Results:

• The proposed model successfully removes noise from OCT images, preserving essential structural details.
• Retinal layer boundaries are accurately extracted as a distinct edge component.
• Experimental results demonstrate superior performance compared to existing speckle noise reduction techniques.
• The developed algorithm significantly improves computational efficiency.

Conclusions:

• The novel variational decomposition model offers an effective solution for noise reduction in OCT imaging.
• Accurate retinal layer segmentation is achieved, facilitating improved clinical diagnosis.
• The method shows significant potential for advancing automated OCT image analysis in ophthalmology.