The abstract is read at the publisher; the summary is JClub's.
A constrained geometric framework for single-input PSOCT improves local optic axis estimation stability under noisy, low-SNR conditions, enabling more reliable depth-resolved fiber orientation in ex vivo human brain tissue and phantoms.
- Why it matters: Conventional single-input PSOCT estimates cumulative retardance and apparent optic axis, which becomes inaccurate when fibers cross over depth, and high noise limits trustworthy local orientation estimation.
- What they did: The method uses polarization super-pixels to cluster spatially connected pixels with similar polarization states, boosting effective SNR and stabilizing recursive matrix inversion, while enforcing a physical constraint from the mirror state of the input polarization during geometric fitting.
- The result: Experiments in ex vivo human brain tissue and phantoms show substantially improved stability and robustness of local optic axis estimation, supporting reliable depth-resolved fiber orientation mapping in fiber-crossing scenarios.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.