Phantom study to determine optimal PET reconstruction parameters for PET/MR imaging of 90Y microspheres following radioembolization. (4th February 2016)

• Record Type: Journal Article
• Authors: Maughan, Nichole M; Eldib, Mootaz; Conti, Maurizio; Knešaurek, Karin; Faul, David; Parikh, Parag J; Fayad, Zahi A; Laforest, Richard
• Is Part Of: Biomedical physics & engineering express. Volume 2:Number 1(2016)
• Publication Date: 2016-02-04
• Subjects: radioembolization -- 90Y -- PET/MRI; Medical physics -- Periodicals; Biophysics -- Periodicals; Biomedical engineering -- Periodicals; Medical sciences -- Periodicals; 610.153
• DOI: 10.1088/2057-1976/2/1/015009
• Languages: English
• ISSNs: 2057-1976

Abstract

Abstract: Yttrium-90 ( 90 Y) radioembolization involves the intra-arterial delivery of radioactive microspheres for treating liver tumors. Although primarily a β − emitter, 90 Y occasionally emits positrons via internal pair production. There is growing interest in positron emission tomography (PET) imaging of these microspheres following radioembolization to localize and quantify their distribution. Due to the low abundance of positron emission (0.0032%), the PET images are highly sensitive to the parameters of iterative reconstruction algorithms used in clinical PET imaging. In this study, we aim to determine the optimal 90 Y PET reconstruction parameters for PET/MRI that provide maximal activity recovery while maintaining a moderate noise level. We imaged an ACR phantom filled with 90 Y chloride solution on two Siemens Biograph mMR scanners with a total activity of 1.00 GBq–7.43 GBq. Images were reconstructed with varying iteration number, post-reconstruction filter size, and scatter correction method. In addition, we evaluated the effects of incorporating point spread function (PSF) compensation as well as reducing the scan time and imaging off the central axis of the scanner in order to simulate realistic patient acquisition protocols as closely as possible. Recovery coefficients (RCs) for hot cylinders 8–25 mm in diameter reached near convergence with limited noise at 3 iterations, 21 subsets, 5 mm FWHM Gaussian post-reconstruction filter, use of PSF compensation, and absolute scatter correction across all scans. RCs increased by as much as 17% and noise decreased by as much as 29% when scan time duration was 30 min compared to only 15 min. Images acquired with the phantom placed 7–10 cm off the central axis of the scanner exhibited slightly reduced RCs, especially for the lowest activity scan. Although optimal reconstruction parameters with current reconstruction software were determined in this study, these parameters failed to achieve perfect count recovery, suggesting further investigation of 90 Y PET imaging is required.