UNLABELLED: PET imaging of D(2) receptors or (18)F-L-dopa metabolism are reference protocols to follow and study neurodegenerative diseases, but the accuracy of striatal PET imaging is limited by the partial-volume effect (PVE). For such studies, the geometric transfer matrix (GTM) method has been proposed to correct the regional mean values for PVE and is now widely used. METHODS: The GTM method models the geometric interactions induced by the PET system between the anatomic regions in which PVE correction is performed. This implies estimation of the corresponding regional spread function (RSF). The literature describes 2 implementations for the RSF calculation; they differ in the way the point spread function (PSF) of the imaging system is modeled, but no comparison or discussion has been given. The first and reference implementation uses an accurate intrinsic detector PSF that is applied in the sinogram space. The second uses a global PSF that is applied in the image space. In this work, we compared the 2 GTM implementations for 3-dimensional (3D) PET striatal imaging using Monte Carlo simulations and a phantom study. We studied the robustness of the GTM correction with respect to residual registration errors between PET and anatomy and with respect to residual segmentation errors. RESULTS: Despite the differences in RSF calculation and computation cost between the 2 implementations, similar recovery results were obtained (between 95% and 100%). The study of robustness of the GTM correction yielded 2 results. A realistic residual misregistration between the anatomic and PET images did not modify the algorithm accuracy but decreased its precision. Conversely, a realistic residual missegmentation of the anatomic regions submitted to GTM correction decreased the correction accuracy. CONCLUSION: A simple but efficient implementation in the image space of the GTM method yields accurate PVE correction in striatal regions in studies with 3D PET and enables clinical use. The method is less sensitive to residual misregistration errors between PET and anatomy than to residual missegmentation of the anatomy. Special care should be taken with segmentation of the regions to correct for PVE.
UNLABELLED: PET imaging of D(2) receptors or (18)F-L-dopa metabolism are reference protocols to follow and study neurodegenerative diseases, but the accuracy of striatal PET imaging is limited by the partial-volume effect (PVE). For such studies, the geometric transfer matrix (GTM) method has been proposed to correct the regional mean values for PVE and is now widely used. METHODS: The GTM method models the geometric interactions induced by the PET system between the anatomic regions in which PVE correction is performed. This implies estimation of the corresponding regional spread function (RSF). The literature describes 2 implementations for the RSF calculation; they differ in the way the point spread function (PSF) of the imaging system is modeled, but no comparison or discussion has been given. The first and reference implementation uses an accurate intrinsic detector PSF that is applied in the sinogram space. The second uses a global PSF that is applied in the image space. In this work, we compared the 2 GTM implementations for 3-dimensional (3D) PET striatal imaging using Monte Carlo simulations and a phantom study. We studied the robustness of the GTM correction with respect to residual registration errors between PET and anatomy and with respect to residual segmentation errors. RESULTS: Despite the differences in RSF calculation and computation cost between the 2 implementations, similar recovery results were obtained (between 95% and 100%). The study of robustness of the GTM correction yielded 2 results. A realistic residual misregistration between the anatomic and PET images did not modify the algorithm accuracy but decreased its precision. Conversely, a realistic residual missegmentation of the anatomic regions submitted to GTM correction decreased the correction accuracy. CONCLUSION: A simple but efficient implementation in the image space of the GTM method yields accurate PVE correction in striatal regions in studies with 3D PET and enables clinical use. The method is less sensitive to residual misregistration errors between PET and anatomy than to residual missegmentation of the anatomy. Special care should be taken with segmentation of the regions to correct for PVE.
Authors: Hui Liu; Chung Chan; Yariv Grobshtein; Tianyu Ma; Yaqiang Liu; Shi Wang; Mitchel R Stacy; Albert J Sinusas; Chi Liu Journal: Phys Med Biol Date: 2015-08-21 Impact factor: 3.609
Authors: C Haegelen; M Verin; B Aubert Broche; F Prigent; P Jannin; B Gibaud; X Morandi Journal: Surg Radiol Anat Date: 2005-09-14 Impact factor: 1.246
Authors: Diego Castillo-Barnes; Li Su; Javier Ramírez; Diego Salas-Gonzalez; Francisco J Martinez-Murcia; Ignacio A Illan; Fermin Segovia; Andres Ortiz; Carlos Cruchaga; Martin R Farlow; Chengjie Xiong; Neil R Graff-Radford; Peter R Schofield; Colin L Masters; Stephen Salloway; Mathias Jucker; Hiroshi Mori; Johannes Levin; Juan M Gorriz Journal: Inf Fusion Date: 2020-01-07 Impact factor: 12.975
Authors: Yi Su; Tyler M Blazey; Abraham Z Snyder; Marcus E Raichle; Daniel S Marcus; Beau M Ances; Randall J Bateman; Nigel J Cairns; Patricia Aldea; Lisa Cash; Jon J Christensen; Karl Friedrichsen; Russ C Hornbeck; Angela M Farrar; Christopher J Owen; Richard Mayeux; Adam M Brickman; William Klunk; Julie C Price; Paul M Thompson; Bernadino Ghetti; Andrew J Saykin; Reisa A Sperling; Keith A Johnson; Peter R Schofield; Virginia Buckles; John C Morris; Tammie L S Benzinger Journal: Neuroimage Date: 2014-12-05 Impact factor: 6.556
Authors: Nikie J Hoetjes; Floris H P van Velden; Otto S Hoekstra; Corneline J Hoekstra; Nanda C Krak; Adriaan A Lammertsma; Ronald Boellaard Journal: Eur J Nucl Med Mol Imaging Date: 2010-04-27 Impact factor: 9.236