Motion artifact and background noise suppression on optical microangiography frames using a naïve Bayes mask

Area of Science:

• Biomedical Imaging
• Optical Coherence Tomography
• Microangiography

Background:

• Optical Coherence Tomography (OCT) enables high-resolution 3D imaging of tissues.
• Optical Microangiography (OMAG) extracts vasculature from OCT data.
• OMAG image quality is degraded by noise and motion artifacts, impacting vessel visualization.

Purpose of the Study:

• To develop a binary mask to reduce background noise in OMAG images.
• To improve the signal-to-noise ratio for better vessel extraction.
• To enable accurate 3D quantification of tissue vasculature.

Main Methods:

• A Naïve Bayes (NB) classification algorithm was trained using expert-segmented images.
• A binary mask was created to filter noise from cross-sectional B-frame OMAG images.
• Results were compared to a frequency rejection filter (FRF) method.

Main Results:

• The NB mask effectively reduced background noise while preserving vessel signals.
• Masked OMAG images showed improved contrast for binarization and quantification.
• NB and FRF methods yielded similar vessel length fractions.
• The NB method is applicable in 3D and not limited to periodic motion.

Conclusions:

• The NB-based binary mask is an effective tool for noise reduction in OMAG imaging.
• This method enhances the visualization and quantification of capillary networks.
• The NB approach offers advantages over FRF for 3D analysis and diverse motion artifacts.