Answers are grounded in this paper and cite their sources.
Polarization-sensitive optical coherence tomography (PSOCT) enables label-free, depth-resolved microstructure mapping in biological tissues, making it a powerful tool for studying myelinated fiber structures in the brain. Conventional single-input PSOCT measures cumulative retardance and apparent optic axis, which become inaccurate when fibers cross over depth. In single-input PSOCT systems, local optic axis estimation is challenging due to high noise levels, limiting the reliability of depth-resolved orientation estimation. In this work, we present a robust framework for local optic axis estimation in single-input PSOCT that incorporates physical constraints and polarization super-pixels. Polarization super-pixels serve as a preprocessing step to cluster spatially connected pixels with similar polarization states, increasing the effective signal-to-noise ratio (SNR) and reducing instability in recursive matrix inversion. In addition, a physical constraint based on the mirror state of the input polarization is enforced during geometric fitting, significantly improving robustness and accuracy under noisy and low-SNR conditions. Experimental validation on ex vivo human brain tissue and phantoms demonstrates substantially improved stability and robustness of local optic axis estimation. This work enables reliable depth-resolved fiber orientation mapping in sophisticated fiber crossing scenarios and extends the potential of single-input PSOCT for analyzing structural connectivity at high depth resolution.
No takes yet. Share an insight, caveat, or question.
Wan et al. (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: