Abstract
Estimation of attenuation from positron emission tomography (PET) data only is of interest for hybrid PET-MR and systems where CT is not available or recommended. However, when using data from a single energy window, emission-based non-time-of-flight (TOF) PET attenuation correction (AC) methods suffer from “cross-talk” artifacts. Based on earlier work, this article explores the hypothesis that cross-talk can be reduced by using more than one energy window. We propose an algorithm for the simultaneous estimation of both activity and attenuation images, as well as, the scatter component of the measured data from a PET acquisition, using multiple energy windows. The model for the measurements is 3-D and accounts for the finite energy resolution of PET detectors; it is restricted to single scatter. The proposed energy-based simultaneous maximum likelihood reconstruction of activity and attenuation with photopeak scatter re-estimation algorithm is compared with simultaneous estimation from a single energy window simultaneous maximum likelihood reconstruction of activity and attenuation with photopeak scatter re-estimation. The evaluation is based on simulations using the characteristics of the Siemens mMR scanner. Phantoms of different complexity were investigated. In particular, a 3-D XCAT torso phantom was used to assess the inpainting of attenuation values within the lung region. Results show that the cross-talk present in non-TOF maximum likelihood reconstruction of activity and attenuation reconstructions is significantly reduced when using multiple energy windows and indicate that the proposed approach warrants further investigation.
Highlights
Results showed that Maximum Likelihood reconstruction of Activity and Attenuation (MLAA) and LBFGS-attenuation correction (AC) achieved the worst and best results, respectively, amongst all the four reconstruction methods
MLAA showed a higher bias in both the estimated attenuation (Fig. 6a) and activity (Fig. 6e) images, whilst both MLAA-EB-S and MLAA-S converged in mean to a similar solution (Fig. 6b-d, Fig. 6f-h), all showing a higher noise level in the lung region compared to the one from an LBFGS-AC (Fig. 6i)
MLAA-EB-S achieved a lower variance with respect to MLAA-S (WW) and MLAA-S (UUstd) and MLAA for both the attenuation (Fig. 6j-m) and activity distributions (Fig. 6n-r)
Summary
Alexandre Bousse was with the Institute of Nuclear Medicine, University College London, London NW1 2BU, UK and is with the Laboratoire de Traitement de l’Information Medicale - LATIM, INSERM, U1101, Universitede Bretagne Occidentale, Brest, France
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: IEEE Transactions on Radiation and Plasma Medical Sciences
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
