Reconstruction for free-space fluorescence tomography using a novel hybrid adaptive finite element algorithm

Area of Science:

• Biomedical Imaging
• Optical Imaging
• Computational Biology

Background:

• Advancements in in-vivo free-space fluorescence molecular imaging and multi-modality imaging for small animals necessitate improved reconstruction methods.
• Current methods face challenges with large datasets and complex, real animal-shape models.

Purpose of the Study:

• To develop and validate a novel hybrid adaptive finite element algorithm for fluorescence tomography reconstruction.
• To address the need for efficient and accurate reconstruction of large datasets in small animal imaging.

Main Methods:

• A linear scheme-based hybrid adaptive finite element algorithm was developed.
• Two inversion strategies, Conjugate Gradient and Landweber iterations, were applied to different mesh levels.
• The algorithm was validated using numerical simulations of a 3-D mouse atlas with a 360-degree free-space fluorescence tomography setup.

Main Results:

• The algorithm demonstrated high computational efficiency.
• High spatial resolution was achieved in the reconstructed models.
• Effective reconstruction was shown for models with irregular shapes and inhomogeneous optical properties.

Conclusions:

• The novel hybrid adaptive finite element algorithm is effective for fluorescence tomography reconstruction in complex small animal models.
• This method offers a significant improvement in computational efficiency and spatial resolution for in-vivo imaging applications.