An Efficient Approach to Perform MR-Assisted PET Data Optimization in Simultaneous PET/MR Neuroimaging Studies

• DOI: 10.2967/jnumed.117.207142
• Publication Date: 2018-06-22T13:00:35Z
• Authors: Kevin T. Chen; Stephanie Salcedo; Kuang Gong; Daniel B. Chonde; David Izquierdo-G...; Alexander Drzezga; Bruce Rosen; Jinyi Qi; Bradford C. Dicke...; Ciprian Catana
• Keywords: Correction for attenuation; Partial volume

A main advantage of PET is that it provides quantitative measures the radiotracer concentration, but its accuracy confounded by factors including attenuation, subject motion, and limited spatial resolution. Using information from one simultaneously acquired morphologic MR sequence with embedded navigators for motion correction (MC), we propose an efficient method, MR-assisted data optimization (MaPET), attenuation (AC), MC, anatomy-aided reconstruction. <b>Methods:</b> For AC, voxelwise linear coefficient maps were generated using SPM8-based method on volume. The used to derive head estimates event-based MC. anatomy provided volume was incorporated into image reconstruction a kernel-based method. Region-based analyses performed assess quality images through various stages optimization. <b>Results:</b> optimized reconstructed MaPET superior in only high signal-to-noise ratio low variation (5.08 0.229 composite cortical region compared 3.12 0.570, <i>P</i> &lt; 10⁻⁴ both comparisons). also shown Cohen's d metric achieve greater effect size distinguishing regions hypometabolism preserved metabolism. <b>Conclusion:</b> We have spatiotemporally correlated single can be partial-volume effects corrections may enable more accurate assessment pathologic changes dementia other brain disorders.