An attenuation measurement technique for rotating planar detector positron tomographs

Area of Science:

• Medical Imaging
• Nuclear Physics
• Positron Emission Tomography (PET)

Background:

• Accurate attenuation correction is crucial for quantitative PET imaging.
• Existing methods face challenges with scattered and accidental coincidences, especially in systems like the Birmingham positron camera.
• Noise amplification during background subtraction limits quantitative accuracy with limited statistics.

Purpose of the Study:

• To develop a new attenuation measurement technique for rotating planar detector PET systems.
• To overcome limitations of traditional attenuation correction methods, particularly noise and scatter.
• To provide quantitatively accurate attenuation correction factors for PET emission images.

Main Methods:

• Utilizing two unshielded positron-emitting line sources attached to detector faces for transmission measurements.
• Rejecting scattered and accidental coincidences by selecting events forming vectors near line sources.
• Reconstructing an attenuation image from measured projections and segmenting it to reduce noise and restore coefficients.

Main Results:

• The proposed technique reconstructs attenuation images with correct features, even with limited statistics.
• Segmentation and renormalization of the attenuation image effectively reduce noise and restore accurate linear attenuation coefficients.
• Reprojection through the segmented image yields quantitatively correct attenuation correction factors.

Conclusions:

• The new technique offers a viable solution for attenuation measurement in PET, especially for systems with limited statistics.
• Reconstruction and segmentation of attenuation images provide a robust method for accurate attenuation correction.
• This approach enhances the quantitative accuracy and statistical quality of PET emission images.