Advancing Localization Optoacoustic Tomography Using Experimentally‐Derived Effective Point Spread Functions

Le résumé fourni par la source

Abstract Localization optoacoustic tomography (LOT) has recently been introduced to overcome the long‐standing resolution barrier imposed by acoustic diffraction in deep‐tissue optoacoustic (OA, photoacoustic) imaging, unlocking new possibilities for microangiographic imaging further enhanced with previously inaccessible functional blood velocity and oxygen saturation readings. The achievable resolution of LOT is intrinsically linked to the precision with which microparticles flowing in blood can be localized, which underscores the need for advanced methods and thorough evaluation of the localization accuracy to optimize imaging performance. Herein, a novel approach is introduced that employs least‐squares fitting of pixel intensities in singular value decomposition (SVD)‐filtered OA images of particles to the experimentally characterized spatially‐inhomogeneous effective point spread functions (ePSFs). Phantom experiments reveal that the proposed approach significantly increases sub‐pixel localization accuracy, with residual errors up to an order of magnitude lower than standard centroid‐ and fitting‐based methods, particular for scenarios where the ePSFs diameters are smaller than the pixel size. In vivo LOT transcranial microangiography of the murine cortex further demonstrates a 13.6% increase in the localized particle counts, translating to a measurable resolution enhancement. As a universal framework, the proposed approach can seamlessly be adapted to other particle‐localization‐based techniques, paving the way for new frontiers in super‐resolution angiography.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.