Enhanced spatial resolution in fluorescence molecular tomography using restarted L1-regularized nonlinear conjugate...

Area of Science:

• Biomedical Imaging
• Optical Imaging
• Medical Physics

Background:

• Fluorescence molecular tomography (FMT) suffers from low spatial resolution due to light scattering and ill-posed inverse problems.
• Traditional L2 regularization methods struggle to preserve details and effectively reduce noise in FMT reconstructions.
• Improving spatial resolution is crucial for accurate molecular imaging in FMT.

Purpose of the Study:

• To develop and evaluate a novel algorithm for enhancing spatial resolution in FMT.
• To investigate the effectiveness of L1 regularization combined with a restarted nonlinear conjugate gradient method for FMT.
• To demonstrate the capability of the proposed algorithm in resolving small, closely spaced fluorescent targets.

Main Methods:

• Developed a restarted L1 regularization-based nonlinear conjugate gradient (re-L1-NCG) algorithm for FMT reconstruction.
• The algorithm employs inner iterations for L1-regularized reconstruction and outer iterations with a restarted strategy to accelerate convergence.
• Validated the algorithm's performance using simulations and physical phantom studies with varying target configurations.

Main Results:

• The re-L1-NCG algorithm demonstrated superior performance in preserving details and reducing noise compared to standard methods.
• Successfully resolved fluorescent targets with an edge-to-edge distance of 0.1 cm at a depth of 1.5 cm.
• Achieved significant improvements in spatial resolution for FMT reconstructions.

Conclusions:

• The re-L1-NCG algorithm offers a computationally efficient and memory-sparing approach for FMT.
• This method provides a substantial advancement in FMT's ability to resolve fine structures, enabling more precise molecular imaging.
• The improved spatial resolution has significant implications for preclinical and clinical applications of FMT.