Optical Coherence Tomography with Gapped Spectrum Using Sparse Iterative Covariance-Based Estimation

Area of Science:

• Biomedical Optics
• Medical Imaging
• Signal Processing

Background:

• Optical coherence tomography (OCT) provides noninvasive, high-resolution cross-sectional imaging of biological tissues.
• Fourier-domain OCT's axial resolution depends on light source spectral bandwidth; gaps degrade resolution and cause artifacts.
• Standard interpolation methods struggle with discontinuous spectra, leading to sidelobes and reduced image quality in OCT.

Purpose of the Study:

• To propose and evaluate a sparse parameter estimation approach, SPICE, for OCT image reconstruction from gapped spectra.
• To enhance axial resolution and suppress sidelobe artifacts in OCT images caused by spectral gaps.
• To compare SPICE's performance against existing methods like GAPES in terms of image quality and computational cost.

Main Methods:

• Developed Sparse Iterative Covariance-based Estimation (SPICE), a sparse parameter estimation framework.
• Applied SPICE to directly resolve depth-dependent components from discontinuous interferograms in OCT data.
• Conducted experiments on multi-layered tape, oral epithelium, and finger skin samples with representative spectral gaps.

Main Results:

• SPICE visually suppressed gap-induced sidelobe artifacts and improved structural interpretability in OCT images.
• Quantitative analysis showed SPICE reduced axial Full Width at Half Maximum (FWHM) by 30-45% and increased Structural Similarity Index Measure (SSIM) by 0.15-0.25.
• SPICE demonstrated significantly lower computational cost compared to GAPES (p < 0.01).

Conclusions:

• SPICE effectively enhances axial resolution and suppresses artifacts in OCT imaging from gapped spectra.
• The method improves image quality and structural interpretability for noninvasive biological tissue imaging.
• SPICE offers a computationally efficient and superior alternative to standard interpolation techniques for OCT reconstruction.