High-speed, sparse-sampling three-dimensional photoacoustic computed tomography in vivo based on principal component...

Area of Science:

• Biomedical Imaging
• Medical Physics
• Computational Imaging

Background:

• Photoacoustic computed tomography (PACT) is a promising multiscale biomedical imaging technology.
• Current PACT systems require high imaging speed, cost-effectiveness, and manageable data flow for clinical translation.
• Iterative reconstruction algorithms, while effective, present computational challenges and long reconstruction times.

Purpose of the Study:

• To develop a novel, rapid, and computationally efficient PACT reconstruction method.
• To enable high-quality 3-D PACT imaging from sparsely sampled data without iterative reconstruction.
• To validate the performance of the new method using in vivo human hand vasculature imaging.

Main Methods:

• Development of a Principal Component Analysis-based PACT (PCA-PACT) reconstruction algorithm.
• Utilizing sparse data sampling for image acquisition.
• Comparison of PCA-PACT with traditional Back-Projection (BP) and compressed sensing-based PACT methods.

Main Results:

• PCA-PACT achieved rapid 3-D image reconstruction from sparse data without iterative processing.
• Compared to BP, PCA-PACT used 50% fewer measurements and 40% less reconstruction time with comparable image quality.
• PCA-PACT demonstrated a sevenfold increase in imaging speed and higher accuracy compared to compressed sensing-based PACT.

Conclusions:

• PCA-PACT offers a significant advancement for rapid, low-cost, 3-D biomedical imaging.
• The method addresses key challenges in PACT system development, including speed and data management.
• PCA-PACT presents a promising approach for diverse preclinical and clinical biomedical applications.