Laplacian manifold regularization method for fluorescence molecular tomography.

Area of Science:

• Biomedical Imaging
• Computational Imaging
• Medical Physics

Background:

• Sparse regularization is crucial for stable 3D reconstruction in fluorescence molecular tomography (FMT).
• Existing L1-regularization methods leverage target distribution sparsity but often overlook spatial structure.
• Integrating spatial information alongside sparsity is key to improving FMT reconstruction.

Purpose of the Study:

• To propose a novel joint L1 and Laplacian manifold regularization model for FMT.
• To enhance the 3D reconstruction performance by incorporating spatial structure information.
• To develop efficient algorithms for solving the proposed regularization model.

Main Methods:

• Developed a joint L1 and Laplacian manifold regularization model for FMT.
• Presented two algorithms, with and without the Barzilai–Borwein strategy, to solve the model.
• Employed numerical simulations and in vivo experiments for validation.

Main Results:

• The proposed Gradient projection-resolved Laplacian manifold regularization method demonstrated superior performance.
• The joint model significantly improved both spatial aggregation and location accuracy.
• Performance was validated against a comparative L1 minimization method.

Conclusions:

• The joint L1 and Laplacian manifold regularization model effectively enhances FMT reconstruction.
• Exploiting both sparsity and spatial structure information leads to more accurate imaging.
• The proposed method offers a significant advancement over traditional L1-based techniques.