Binary photoacoustic tomography for improved vasculature imaging.

Area of Science:

• Medical Imaging
• Biomedical Engineering
• Computational Imaging

Background:

• Photoacoustic tomography (PAT) is crucial for reconstructing vascular networks in cancer research, cardiovascular studies, and neuroimaging.
• Current PAT methods often involve a two-step process of image reconstruction followed by segmentation.
• Direct reconstruction of vascular networks remains a significant challenge in photoacoustic imaging.

Purpose of the Study:

• To introduce and evaluate a novel binary photoacoustic tomography (PAT) approach for direct vascular network reconstruction.
• To demonstrate the capability of binary PAT to recover vasculature structures accurately from photoacoustic sinogram data.
• To compare the performance of binary PAT against established reconstruction methods.

Main Methods:

• Developed a binary tomography algorithm that solves a dual-optimization problem for direct image reconstruction.
• Images are reconstructed with each pixel yielding a binary outcome (background or absorber).
• Compared the binary tomography approach against backprojection, Tikhonov regularization, and sparse recovery methods.

Main Results:

• The binary tomography approach demonstrated superior vasculature recovery compared to other methods.
• Achieved a 10% improvement in vasculature recovery using in-silico data, measured by the Dice similarity coefficient.
• Validated performance through numerical simulations, physical phantom experiments, and in-vivo rat brain vasculature data.

Conclusions:

• The proposed binary tomography algorithm accurately recovers vasculature structures, even with limited data.
• This approach shows potential for applications in therapy monitoring and hemorrhage detection.
• Binary PAT offers a promising advancement for precise vascular network imaging.