18F, 68Ga, 89Zr Accreditation specifications
The EARL accreditation is based on quality control (QC) experiments that were described and published in the EANM guidelines in 2010. Meanwhile the field has moved forward and new developments, technologies and insights have become available. During the EANM Conference 2018 in Düsseldorf EARL announced the update of the EARL performance specifications. 18F standards 2 allows sites to benefit from state-of-the-art PET/CT technology and is based on a recent publication by EARL entitled “Feasibility of state-of-the-art PET/CT systems performance harmonisation”. The impact of the new 18F standard 2 on the quantification of clinical studies has been described in “Quantitative implications of the updated EARL 2019 PET–CT performance standards”. EARL is dedicated towards improving and supporting the quality of nuclear medicine procedures. You can review the presentation give by Prof. Dr. R. Boellaard in regards to this EARL update here.
Since 2026, EARL 18F standards 1 and 2 are transitioning from SUV recovery coefficients (RCs) to contrast recovery coefficients (CRCs) in PET scanner accreditation in order to create a more standardized, efficient, and consistent approach across clinical trials globally. The changes are based on unified work done and recently published: Standardization of PET/CT Performance Requirements for Whole-Body Quantitative Imaging: An International Proposal.
During EANM’26 EARL released a new 18F standards 3, which should be considered as prototype specifications. The newly proposed standards 3 will go through a field-testing phase during the first year (or when sufficient data has been collected) and may be refined slightly depending on the results from a larger set of real -world data. Therefore, the current EARL 18F standards 2 are still the recommended standards for any type of scanner to be used for clinical or research studies. The prototype EARL 18F standards 3 are specifically intended for ultra-high sensitivity PET systems, which are typically PET/CT systems with a longer axial field of view (e.g. typically 50cm or larger, although any high sensitivity system may apply for EARL standards 3). EARL standards 3 will generate images with a high spatial resolution, which require sufficient count statistics to assure minimal SUV bias, in particular for SUVmax.
Calibration QC acceptability criteria for 18F, 68Ga and 89Zr are set to +/-10% for each isotope independently.
18F standards 3 effective as of 01 November 2026; CRCs effective as of quarter 1 of 2026; SUV recovery coefficients effective as of quarter 3 of 2020:
NEMA IEC phantom spheres | 18F standards 1 CRCs** 18F standards 1 SUV RCs | 18F standards 2 CRCs** 18F standards 2 SUV RCs | 18F standards 3 CRCs** 18F standards 3 SUV RCs | ||||||
Diameter (mm) | Volume (mL) | max | mean | max* | mean | peak** | max* | mean | peak** |
37 | 26.52 | 0.94 – 1.18 0.95 – 1.16 | 0.73 – 0.87 0.76 – 0.89 | 1.06 – 1.33 1.05 – 1.29 | 0.83 – 1.00 0.85 – 1.00 | 0.89 – 1.11 0.90 – 1.10 | 1.06 – 1.33 1.05 – 1.29 | 0.83 – 1.00 0.85 – 1.00 | 0.89 – 1.11 0.90 – 1.10 |
28 | 11.49 | 0.90 – 1.15 0.91 – 1.13 | 0.68 – 0.83 0.72 – 0.85 | 1.01 – 1.30 1.01 – 1.26 | 0.79 – 0.97 0.82 – 0.97 | 0.89– 1.11 0.90 – 1.10 | 1.06 – 1.33 1.05 – 1.29 | 0.83 – 1.00 0.85 – 1.00 | 0.89 – 1.11 0.90 – 1.10 |
22 | 5.57 | 0.81 – 1.10 0.83 – 1.09 | 0.58 – 0.75 0.63 – 0.78 | 1.01 – 1.37 1.01 – 1.32 | 0.77– 0.99 0.80 – 0.99 | 0.89 – 1.11 0.90 – 1.10 | 1.06 – 1.33 1.05 – 1.29 | 0.83 – 1.00 0.85 – 1.00 | 0.89 – 1.11 0.90 – 1.10 |
17 | 2.57 | 0.69 – 1.01 0.73 – 1.01 | 0.51– 0.69 0.57 – 0.73 | 1.00 – 1.43 1.00 – 1.38 | 0.73 – 0.97 0.76 – 0.97 | 0.71 – 0.99 0.75 – 0.99 | 1.06 – 1.33 1.05 – 1.29 | 0.83 – 1.00 0.85 – 1.00 | 0.89 – 1.11 0.90 – 1.10 |
13 | 1.15 | 0.53 – 0.83 0.59 – 0.85 | 0.36 – 0.54 0.44 – 0.60 | 0.83 – 1.25 0.85 – 1.22 | 0.58 – 0.84 0.63 – 0.86 | 0.37 – 0.66 0.45 – 0.70 | 1.16 – 1.43 1.14 – 1.38 | 0.83 – 1.00 0.85 – 1.00 | 0.60 – 0.83 0.65 – 0.85 |
10 | 0.52 | 0.25 – 0.51 0.34 – 0.57 | 0.17 – 0.35 0.27 – 0.43 | 0.45 – 0.86 0.52 – 0.88 | 0.30 – 0.55 0.39 – 0.61 | 0.17 – 0.33 0.27 – 0.41 | 1.06 – 1.33 1.05 – 1.29 | 0.71 – 0.94 0.75 – 0.95 | 0.26 – 0.49 0.35 – 0.55 |
*SUVmax for total body PET/CT scanners is under investigation
**SUVpeak and CRCs limits are under a revision
Recovery coefficient values are calculated based on:
- RCSUVmean – 50% background corrected isocontour VOI
- RCSUVmax – the maximum voxel value included in VOI
- RCSUVpeak – spherical VOI with a diameter of 12 mm, positioned so to yield the highest uptake
VOIs used for the analysis are based on the PET data, not on the CT.
EARL FDG PET/CT optimization procedure: EARL procedure for assessing PET/CT system specific patient FDG activity preparations for quantitative FDG PET/CT studies. R.Boellaard, A.T. Willemsen, B.Arends, E.P. Visse