Impact of reconstruction algorithms at different sphere-to-background ratios on PET quantification: A phantom study.

Applied radiation and isotopes : including data, instrumentation and methods for use in agriculture, industry and medicine [Appl Radiat Isot] 2025 Jun; Vol. 220, pp. 111761. Date of Electronic Publication: 2025 Feb 27.

Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Using National Electrical Manufacturers Association (NEMA) phantom, the behavior of four distinct Positron Emission Tomography/Computed Tomography (PET/CT) reconstruction algorithms was investigated. These reconstruction algorithms were (Ordered Subset Expectation Maximization (OSEM), OSEM+ (Point Spread Function) PSF, OSEM + Time of Flight (TOF), and OSEM + TOF + PSF), and the focus was on sphere sizes and SBRs using recovery coefficients as a quantitation method. The obtained results demonstrated the significant effect of TOF on Gibbs artifact and Partial Volume Effect (PVE) at various Sphere-to-Background Ratios (SBRs). TOF-based algorithms improved quantification accuracy and mitigated the influence of Gibbs artifact, particularly at higher SBRs. Compared to PSF algorithm, TOF- based algorithms effectively mitigated the impact of PVE on small-sized spheres and less dependent on SBRs. In terms of Standardized Uptake Value (SUV) quantification, SUVmean was better when utilizing TOF-based algorithms at lower SBRs, whereas SUVmax at higher SBRs. The combination of TOF and PSF produced a promising outcomes in quantifying and detecting a small-sized spheres across various SBRs, ultimately resulting in a more reliable and precise diagnostic information.
(Copyright © 2025 Elsevier Ltd. All rights reserved.)

English

Publisher: Pergamon Press Country of Publication: England NLM ID: 9306253 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1872-9800 (Electronic) Linking ISSN: 09698043 NLM ISO Abbreviation: Appl Radiat Isot Subsets: MEDLINE

Phantoms, Imaging* , Algorithms* , Positron Emission Tomography Computed Tomography*/methods , Positron Emission Tomography Computed Tomography*/statistics & numerical data , Positron-Emission Tomography* , Image Processing, Computer-Assisted*/methods, Humans ; Artifacts

20250512