A novel approach for quantitative harmonization in PET

Objectives: PET scanner harmonization is recommended by EANM/EARL1 to achieve comparable quantitative performance across multiple sites. However, the task is often beyond the capabilities of small non-academic centers due to the complex scanning and analysis required2. In this work we introduce a novel harmonization methodology based on simple cylindrical phantom measurements and show that it can match the performance of more complex harmonization approaches. Methods: Axial and radial edge spread functions and noise power spectra were measured in cylindrical uniformity phantoms by adapting methods developed for CT imaging3. These measurements were used as inputs to a PET image generation model based on convolution plus a signal-dependent4 and correlated noise term. The model was used to simulate images of the more complex NEMA image quality phantom with spherical inserts. An optimization algorithm5 was used to find the optimal smoothing filters for the simulated NEMA phantom images to identify those that best harmonized the PET scanners. We validated our method by acquiring 18F-FDG uniformity and NEMA phantom scans with 10:1 sphere to background activity concentration ratio2. The measured and simulated contrast recovery coefficients were compared on Bland-Altman plots. Seven different PET models from two manufacturers installed at five institutions were included. Results: Our methodology is able to predict contrast recovery coefficients (CRCs) from NEMA phantoms with errors below ±5.2% for CRCmax and ±3.7% for CRCmean (limits of agreement = 95%, see Figure 1). The prediction bias was 0.67% for CRCmax and 0.36% for CRCmean. Optimal post-filters for harmonization varied between 5.8 and 7.5 mm FWHM. After applying the proposed harmonization protocol, all the CRC values were within the tolerances from EANM1 (see supplemental figure). Conclusions: Measurements of simple PET uniformity phantom can sufficiently characterize a PET system’s image quality and be used to quantitatively harmonize PET scanners. Quantitative harmonization in compliance with the EARL FDG-PET/CT accreditation program is achieved in a simpler way, without the need of NEMA phantoms. This may lead to simplified scanner harmonization workflows more accessible to smaller institutions.