Comparison of PET template-based and MRI-based image processing in the quantitative analysis of C-11-raclopride PET

作者：Kuhn Felix P^*; Warnock Geoffrey I; Burger Cyrill; Ledermann Katharina; Martin Soelch Chantal; Buck Alfred

来源：EJNMMI Research, 2014, 4(1): 7.

DOI：10.1186/2191-219X-4-7

• Neuroimaging
• PET/MRI
• Raclopride
• Movement disorders

Background: Quantitative measures of C-11-raclopride receptor binding can be used as a correlate of postsynaptic D-2 receptor density in the striatum, allowing C-11-raclopride positron emission tomography (PET) to be used for the differentiation of Parkinson%26apos;s disease from atypical parkinsonian syndromes. Comparison with reference values is recommended to establish a reliable diagnosis. A PET template specific to raclopride may facilitate direct computation of parametric maps without the need for an additional MR scan, aiding automated image analysis. %26lt;br%26gt;Methods: Sixteen healthy volunteers underwent a dynamic C-11-raclopride PET and a high-resolution T1-weighted MR scan of the brain. PET data from eight healthy subjects was processed to generate a raclopride-specific PET template normalized to standard space. Subsequently, the data processing based on the PET template was validated against the standard magnetic resonance imaging (MRI)-based method in 8 healthy subjects and 20 patients with suspected parkinsonian syndrome. Semi-quantitative image analysis was performed in Montreal Neurological Institute (MNI) and in original image space (OIS) using VOIs derived from a probabilistic brain atlas previously validated by Hammers et al. (Hum Brain Mapp, 15: 165-174, 2002). %26lt;br%26gt;Results: The striatal-to-cerebellar ratio (SCR) of C-11-raclopride uptake obtained using the PET template was in good agreement with the MRI-based image processing method, yielding a Lin%26apos;s concordance coefficient of 0.87. Bland-Altman analysis showed that all measurements were within the +/- 1.96 standard deviation range. In all 20 patients, the PET template-based processing was successful and manual volume of interest optimization had no further impact on the diagnosis of PD in this patient group. A maximal difference of %26lt;5% was found between the measured SCR in MNI space and OIS. %26lt;br%26gt;Conclusions: The PET template-based method for automated quantification of postsynaptic D-2 receptor density is simple to implement and facilitates rapid, robust and reliable image analysis. There was no significant difference between the SCR values obtained with either PET-or MRI-based image processing. The method presented alleviates the clinical workflow and facilitates automated image analysis.

• 出版日期2014-1-22