Accuracy and precision of partial volume correction in oncological PET/CT studies

Accurate quantification of tracer uptake in small tumors using positron emission tomography (PET) is hampered by partial volume effects as well as by methods of volume of interest (VOI) delineation. This study aimed to investigate the effect of partial volume correction (PVC) combined with several VOI methods on accuracy and precision of quantitative PET.

Four image-based PVC methods and resolution modeling (applied as PVC) were used in combination with several commonly used VOI methods. Performance was evaluated using simulations, phantom experiments, and clinical repeatability studies. Simulations were based on a whole-body (18)F-fluorodeoxyglucose-PET scan in which differently sized spheres were placed in lung and mediastinum. National Electrical Manufacturers Association NU2 Quality phantom was used for the experiments. Repeatability data consisted of a (18)F-fluorodeoxyglucose-PET/CT study in 11 advanced non-small cell lung cancer patients and a (18)F-fluoromethylcholine-PET/CT study in 12 metastatic prostate cancer patients.

Phantom data demonstrated PVC is strongly affected by the applied resolution kernel, with accuracy differing ~20-50% between full-width at half-maximum setting of 5.0 and 7.5mm. For all PVC methods, large differences in accuracy were seen between VOI methods. Additionally, variable sensitivity of image-based PVC methods to volumetric accuracy of VOI methods was observed. For most PVC methods, accuracy was strongly affected by >2.5mm misalignment of simulated true VOI. When using most optimal VOI method per PVC method, high accuracy could be achieved. For example, resolution modeling for mediastinal lesions and iterative deconvolution for lung lesions were 99±1.5% and 99±0.9% accurate, respectively, for spheres ≥15mm in diameter. Precision worsened slightly for resolution modeling, and to a larger extent for some image-based PVC methods. Uncertainties in delineation propagated into uncertainties in PVC performance, as confirmed in clinical data.

Accuracy and precision of PVC methods tested depended strongly on VOI method, resolution settings, contrast, and spatial alignment of VOI. PVC has the potential to improve accuracy of tracer uptake assessment substantially, provided robust and accurate VOI methods become available. Commonly used delineation methods may not be adequate for this purpose.

Journal of nuclear medicine : official publication, Society of Nuclear Medicine. 2016 May 26 [Epub ahead of print]

Matthijs C F Cysouw, Gerbrand Maria Kramer, Otto S Hoekstra, Virginie Frings, Adrianus Johannes de Langen, Egbert F Smit, Alfons J M van den Eertwegh, Daniela E Oprea-Lager, Ronald Boellaard

VU University Medical Center, Netherlands;, VU University Medical Center, Netherlands;, VU University Medical Center, Netherlands;, VU University Medical Center, Netherlands;, VU University Medical Center, Netherlands;, VU University Medical Centre, Netherlands;, VU University Medical Center, Netherlands;, VU University Medical Center, Netherlands;, University Medical Center Groningen, Netherlands.