Efficient image reconstruction for a small animal PET system with dual-layer-offset detector design

Area of Science:

• Medical Imaging
• Biomedical Engineering
• Nuclear Medicine

Background:

• The Inliview-3000B is a compact PET/SPECT/CT system for small animal research.
• Its dual-layer-offset PET detector design presents challenges in normalization and sensitivity map calculations.
• Larger crystal cross-sections (2 mm) in the PET subsystem hinder high-resolution image reconstruction.

Purpose of the Study:

• To develop an efficient image reconstruction method for the Inliview-3000B PET subsystem.
• To achieve high imaging performance, including improved resolution and signal-to-noise ratio.

Main Methods:

• List mode reconstruction with on-the-fly system modeling, including multi-ray tracing for accurate physical process simulation.
• Iterative estimation of crystal efficiency combined with phantom data and simulations for normalization correction.
• Modified data organization to index all useful lines-of-response (LORs) for stacked detector layers.

Main Results:

• The iterative crystal efficiency estimation achieved good accuracy.
• Random crystal sampling in reconstruction demonstrated superior performance with reduced computation compared to uniform sampling.
• The method successfully resolved 1.1 mm hot rods in phantom studies and provided high-resolution, high-SNR images of small animals.

Conclusions:

• Introducing randomness into image reconstruction enhances performance and reliability for high-resolution PET imaging.
• The Inliview-3000B PET subsystem, with the proposed reconstruction, offers detailed insights for preclinical research on small animals.