Information loss and reconstruction in diffuse fluorescence tomography.

Area of Science:

• Biomedical Optics
• Medical Imaging
• Computational Science

Background:

• Diffuse fluorescence tomography (DFT) is a key imaging modality.
• Improving spatial resolution in DFT is crucial for diagnostic accuracy.
• Current methods like Tikhonov regularization and SVD have limitations.

Purpose of the Study:

• To theoretically investigate the impact of mesh density on spatial resolution in DFT.
• To determine if increasing mesh density beyond measurement count enhances optical reconstruction resolution.
• To identify factors limiting spatial resolution improvements in DFT.

Main Methods:

• Theoretical analysis using linear algebra.
• Modeling light transport using the diffusion approximation.
• Evaluating reconstruction techniques including Tikhonov regularization and truncated singular value decomposition.

Main Results:

• Increasing mesh node density beyond the number of measurements does not improve spatial resolution.
• Excessive node density leads to information loss in the forward model.
• Common reconstruction techniques cannot recover this lost information.
• The smoothing properties of the elliptic forward model contribute to resolution limitations.

Conclusions:

• Mesh density is not the primary factor for improving DFT spatial resolution.
• Boundary-sensitive reconstruction methods (e.g., L1-reconstruction, priors) are beneficial.
• Optimizing measurement geometry is essential for achieving desired image resolution.