Improving dOCT image quality with short sequences and automated binning.

Area of Science:

• Biomedical Imaging
• Optical Coherence Tomography
• Signal Processing

Background:

• Dynamic optical coherence tomography (dOCT) leverages signal fluctuations for tissue contrast.
• Motion artifacts are a significant challenge in dOCT, potentially degrading image quality and diagnostic accuracy.
• Optimizing image processing techniques is crucial for enhancing dOCT performance.

Purpose of the Study:

• To investigate the impact of reducing the time base for signal fluctuation evaluation in dOCT.
• To introduce and evaluate an automatic clustering method (neural-gas algorithm) for optimizing color channel borders.
• To quantitatively assess the reduction in motion-induced noise and the maintenance of image quality.

Main Methods:

• Implementation of a shortened time base for signal fluctuation analysis in dOCT.
• Application of the neural-gas algorithm for automatic optimization of color channel boundaries.
• Quantitative evaluation using Mean Squared Error (MSE), Peak Signal to Noise Ratio (PSNR), and Structural Similarity (SSIM) on 15 tissue samples.

Main Results:

• Reducing the time base for signal fluctuation evaluation effectively minimizes motion-induced noise.
• The neural-gas algorithm successfully optimized the border between color channels, enhancing image quality.
• Quantitative metrics confirmed a favorable trade-off between noise reduction and image quality preservation.

Conclusions:

• Shortened time base evaluation is a viable strategy to mitigate motion artifacts in dOCT.
• Automatic border optimization using neural-gas algorithm improves dOCT image quality and robustness.
• This approach enhances the reliability of dOCT for various tissue imaging applications.