Quantitative predictions in small-animal X-ray fluorescence tomography

Area of Science:

• Medical Imaging
• Computational Biology
• Nanotechnology

Background:

• X-ray fluorescence (XRF) tomography using nanoparticles (NPs) offers high-resolution molecular imaging in small animals.
• Accurate reconstruction of NP distribution is crucial but lacks methods for signal prediction and accuracy evaluation in realistic scenarios.

Purpose of the Study:

• To present a GPU-based computational model for small-animal XRF tomography.
• To enable realistic full-body simulations for evaluating quantitative reconstruction algorithms.

Main Methods:

• Developed a GPU-based computational model combining an accelerated Monte Carlo tool with a small-animal phantom.
• Simulated experimental systems to assess reconstruction algorithm performance on organs and tumors.
• Evaluated detection limits for sub-millimeter tumors at realistic NP concentrations.

Main Results:

• The model allows for unprecedented realistic full-body simulations of small-animal XRF tomography.
• Quantitative performance and accuracy of reconstruction algorithms were evaluated for various anatomical structures.
• Detection limits for sub-millimeter tumors were predicted under realistic nanoparticle concentrations.

Conclusions:

• The developed computational model is a valuable tool for optimizing experimental setups and reconstruction algorithms in small-animal XRF tomography.
• This work advances the field by providing a feasible method for predicting signal levels and evaluating reconstruction accuracy.
• Facilitates the development of next-generation molecular imaging techniques using nanoparticles.