Nonlinear greedy sparsity-constrained algorithm for direct reconstruction of fluorescence molecular lifetime...

Area of Science:

• Biomedical optics
• Medical imaging
• Computational modeling

Background:

• Time-domain fluorescence molecular lifetime tomography (TD-FMLT) is crucial for in vivo molecular imaging.
• Improving spatial resolution in TD-FMLT remains a significant challenge.
• Existing methods often struggle with closely spaced targets.

Purpose of the Study:

• To introduce a novel algorithm, accelerated nonlinear orthogonal matching pursuit (ANOMP), for enhanced TD-FMLT.
• To improve the spatial resolution and quantification accuracy of TD-FMLT.
• To validate the algorithm's performance in phantom and in vivo experiments.

Main Methods:

• Developed an ANOMP algorithm, a nonlinear greedy sparsity-constrained method.
• ANOMP employs outer iterations to select elements based on gradient information and inner iterations for intermediate estimates.
• A stability-based stopping criterion ensures robustness across different edge-to-edge distances (EEDs).

Main Results:

• ANOMP successfully improved spatial resolution in TD-FMLT.
• High quantification accuracy was achieved, even with small EEDs.
• Phantom and in vivo mouse experiments confirmed the algorithm's effectiveness.

Conclusions:

• The proposed ANOMP algorithm significantly advances TD-FMLT capabilities.
• ANOMP offers a robust solution for high-resolution molecular lifetime tomography.
• This method holds promise for improved biomedical imaging applications.