Fluorescence molecular tomography based on an online maximum a posteriori estimation algorithm.

Area of Science:

• Biomedical imaging
• Optical molecular imaging
• Medical physics

Background:

• Fluorescence molecular tomography (FMT) is a sensitive, non-invasive imaging technique for early tumor detection.
• FMT reconstruction faces challenges due to limited measurement data and significant light scattering, leading to ill-posed inverse problems.
• Improving FMT reconstruction quality and efficiency is crucial for clinical applications.

Purpose of the Study:

• To enhance Fluorescence Molecular Tomography (FMT) reconstruction quality and efficiency.
• To address the ill-posed nature of the FMT inverse problem.
• To introduce a novel algorithm for faster and more accurate FMT image reconstruction.

Main Methods:

• Developed a reconstruction model incorporating log-sum regularization.
• Introduced an online maximum a posteriori estimation (OPE) algorithm to solve the non-convex optimization problem.
• Evaluated the OPE algorithm's performance using simulations and experimental data.

Main Results:

• The OPE algorithm demonstrated a remarkable speed of convergence.
• Simulations and experiments confirmed the OPE algorithm's ability to ensure good reconstruction quality.
• The proposed method exhibited outstanding performance in terms of reconstruction efficiency.

Conclusions:

• The combination of log-sum regularization and the OPE algorithm significantly improves FMT reconstruction.
• The OPE algorithm offers a computationally efficient solution for FMT, enhancing its practical utility.
• This advancement holds promise for more effective early tumor detection using optical molecular imaging.