Autocorrelation noise removal for optical coherence tomography by sparse filter design.

Area of Science:

• Biomedical Optics
• Signal Processing
• Medical Imaging

Background:

• Autocorrelation noise (ACN) is a significant artifact in optical coherence tomography (OCT) imaging.
• ACN degrades image quality and can obscure underlying sample structures.
• Existing methods for ACN reduction in OCT have limitations.

Purpose of the Study:

• To present a novel reconstruction method for effective elimination of autocorrelation noise (ACN) in optical coherence tomography (OCT).
• To improve the signal-to-noise ratio and clarity of OCT images.

Main Methods:

• The proposed method models optical fields scattered from sample features as a sparse finite impulse response (FIR) filter.
• OCT reconstruction is framed as a parameter identification problem for a sparse FIR filter.
• Parameter identification is achieved using [script-l](1) optimization with soft thresholding.

Main Results:

• Experimental results demonstrate the effective attenuation of autocorrelation noise (ACN).
• The proposed method yields OCT reconstruction results with significantly reduced ACN.
• Improved image quality and feature visualization in OCT are achieved.

Conclusions:

• The presented sparse FIR filter-based reconstruction method is effective in eliminating ACN in OCT.
• This approach offers a promising solution for enhancing OCT image quality.
• The method has the potential to improve diagnostic capabilities in OCT-based applications.