Objective: Quantitative PET requires accurate attenuation correction (AC), which is conventionally achieved using CT-derived maps, incurring extra dose and complexity. This study explores a CT-less AC approach for the plastic scintillator-based J-PET scanner by investigating whether detector-scattered photons (DS-LORs) can serve as an intrinsic transmission source. Methods: • Simulation Setup: The study utilized GATE v9.1 software to simulate the modular J-PET scanner, which consists of plastic scintillator strips where 511 keV photons predominantly undergo Compton scattering. Two configurations were evaluated: a uniform water cylinder and a NEMA IQlike phantom containing a cold lung insert and six hot spheres to simulate heterogeneous activity. • Analysis & reconstruction: The study isolated triple-hit coincidences to analyze angular deviations. DS-LORs were identified using a specific scatter test metric based on interaction positions and detection times. Attenuation maps were reconstructed using Filtered Back Projection (FBP) applied to the ratio of water-to-air sinograms. Results: Due to the properties of plastic scintillators, DS-LORs constituted approximately 50% of all detected prompts, providing a substantial dataset for analysis. In a simulated 10-minute acquisition, the noise propagation analysis yielded a Signal-to-Noise Ratio (SNR) of ≈ 33.9 for water. This is significantly higher than conventional crystal-based PET, which yields only ≈ 5% DS-LORs and a lower SNR. Reconstructions successfully resolved the NEMA phantom features, including the cold lung insert and hot spheres, particularly when phantom-scattered events were rejected. Conclusions and future directions: This study confirms that DS-LORs retain sufficient attenuation information to estimate μ-maps, demonstrating the potential for fully CT-less PET imaging. While the current implementation faces limitations like reliance on blank scans and activity-attenuation crosstalk, the high statistics of DS-LORs are a foundation for developing joint reconstruction algorithms aiming towards clinical application.
No takes yet. Share an insight, caveat, or question.
Tiwari et al. (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: