List-mode image reconstruction for positron emission tomography using tetrahedral voxels

Area of Science:

• Medical Imaging
• Computational Physics
• Voxel-based analysis

Background:

• Tetrahedral voxel decomposition offers advantages for emission tomography, including multi-resolution capabilities and volumetric partitioning.
• Existing methods using tetrahedral decomposition often rely on pre-calculated system matrix elements, limiting flexibility and region-of-interest definition.
• Accurate motion compensation, especially non-rigid deformations, is crucial for awake animal positron emission tomography (PET) and small brain region quantification.

Purpose of the Study:

• To introduce an on-the-fly calculation of system matrix elements using tetrahedral decomposition for emission tomography.
• To demonstrate the application of this novel approach in awake animal PET, addressing challenges in motion compensation and quantification.

Main Methods:

• Image space decomposition using tetrahedral voxels.
• List-mode, on-the-fly calculation of system matrix elements.
• Application to awake animal PET imaging with rigid and non-rigid motion compensation.

Main Results:

• The developed algorithm enables accurate, event-based motion compensation, including non-rigid deformations.
• This method enhances the flexibility of tetrahedral decomposition in emission tomography.
• Facilitates precise quantification within small brain regions in awake animal PET.

Conclusions:

• Tetrahedral voxel decomposition with on-the-fly system matrix calculation provides a flexible and accurate approach for emission tomography.
• This technique is particularly beneficial for motion-corrected awake animal PET imaging.
• Enables precise quantification and analysis in challenging imaging scenarios.