Filters for three-dimensional limited-angle tomography

Area of Science:

• Medical Imaging
• Image Reconstruction
• Positron Emission Tomography

Background:

• Three-dimensional tomographic reconstruction is crucial for analyzing data from stationary positron cameras.
• Frequency-space filtering of back-projected images is a necessary step in this reconstruction process.

Purpose of the Study:

• To derive a closed-form expression for the appropriate frequency-space filter.
• To address the challenge of limited angular projection data inherent in planar dual-detector systems.

Main Methods:

• Derivation of filter expressions in the frequency domain.
• Analysis of data from planar dual-detector stationary positron cameras.
• Consideration of detector cross-section shapes (rectangular and circular).

Main Results:

• A suitable closed-form filter expression has been obtained despite the limited angular range of projections.
• The derived expressions are applicable to both rectangular and circular detector cross-sections.
• This provides a practical tool for improving tomographic reconstruction.

Conclusions:

• The derived frequency-space filter is essential for accurate 3D tomographic reconstruction with stationary positron cameras.
• The method accounts for the practical limitation of restricted projection data.
• This work facilitates enhanced image quality in positron emission tomography.