Three-Dimensional Photoacoustic Tomography with Ultrasound Localization Priors.

Area of Science:

• Biomedical Imaging
• Optical Imaging
• Ultrasound Imaging

Background:

• Three-dimensional photoacoustic tomography (3D-PAT) offers noninvasive imaging but faces limitations in spatial resolution due to acoustic diffraction.
• Incomplete detection geometry in 3D-PAT can degrade image fidelity and quantitative accuracy, particularly for vascular structures.

Purpose of the Study:

• To develop a novel reconstruction framework, 3D-PAULM prior, integrating sub-diffraction vascular information from ultrasound localization microscopy (ULM) into 3D-PAT.
• To improve spatial resolution, image fidelity, and quantitative accuracy of 3D-PAT, especially for vascular imaging and blood oxygen saturation estimation.

Main Methods:

• Developed a ULM-guided, model-based reconstruction framework (3D-PAULM prior) for 3D-PAT.
• Incorporated sub-diffraction vascular priors from ULM using weighted regional Laplacian regularization.
• Validated the framework using numerical simulations, tissue-mimicking phantoms, and in vivo mouse brain imaging.

Main Results:

• 3D-PAULM prior improved spatial resolution by over 50% compared to conventional reconstruction.
• Achieved a 261.2% increase in contrast-to-noise ratio and a 24.6% enhancement in structural similarity index.
• Successfully recovered poorly resolved or missing vascular structures in vivo and produced more accurate blood oxygen saturation maps.

Conclusions:

• 3D-PAULM prior represents a robust multimodal reconstruction strategy for high-resolution structural and functional photoacoustic imaging.
• The integration of ULM priors significantly overcomes the resolution limitations of conventional 3D-PAT.
• This approach holds promise for advancing noninvasive vascular imaging and functional assessment.