Impact of Point-Spread Function Modeling on Image Quality in ⁶⁸ Ga-PSMA PET/CT

The objective of this study was to evaluate the impact of 2 PET reconstruction methods—time of flight (TOF) and TOF with point-spread function (PSF) modeling (TOF + PSF)—on image quality and measured quantitative metrics in ⁶⁸Ga-PSMA PET/CT for prostate cancer. Methods: This retrospective analysis included 36 patients with prostate cancer who underwent ⁶⁸Ga-PSMA PET/CT at the Sultan Qaboos Comprehensive Cancer Care and Research Center between October 2022 and February 2024. PET data were reconstructed using TOF and TOF + PSF on a Biograph Vision 600 system. The quantitative parameters compared were SUVmean, prostate-specific membrane antigen tumor volume (PSMA-TV), total lesion uptake (TLU), signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and lesion-to-background ratio (LBR; lesion SUVmean/background SUVmean). Paired t tests or Wilcoxon signed-rank tests were used as appropriate, and agreement between reconstructions was assessed using Bland–Altman analysis. Results: Compared with TOF, TOF + PSF yielded a significantly higher SUVmean (mean difference, 1.41; 95% CI, 1.08–1.75; P < 0.001), SNR (mean difference, 23.31; 95% CI, 18.34–28.28; P < 0.001), CNR (mean difference, 23.54; 95% CI, 18.62–28.47; P < 0.001), LBR (mean difference, 8.0; 95% CI, 5.85–10.16; P < 0.0001), and TLU (mean difference, 3.62; 95% CI, 2.40–4.84; P < 0.001). PSMA-TV showed no significant difference between reconstructions (mean difference, 0.0052; 95% CI, −0.0154 to 0.0257; P = 0.626). Bland–Altman analysis demonstrated a systematic positive bias for TOF + PSF in SUVmean, SNR, CNR, LBR, and TLU. Because PSF modeling can increase SUV-based measurements via resolution recovery and edge effects, these increases should be interpreted as reconstruction-dependent changes in relative contrast rather than confirmed improvements in absolute quantitative accuracy without ground-truth validation. Conclusion: TOF + PSF reconstruction improves image quality and increases SUV-based and contrast-related metrics in ⁶⁸Ga-PSMA PET/CT and thus may enhance the conspicuity of small or low-contrast lesions. Phantom-based validation is recommended to determine the extent to which these reconstruction-dependent differences translate into improved absolute quantitative accuracy.