. Evaluation of Voxel Concordance-Discordance Between [11C]PIB and Florbetapir PET Group Difference T-Maps for Amyloid(+) Vs. Amyloid(-) Subject Groups. Human Amyloid Imaging Abstract. 2012 Jan 1;


Objective: Apply a non-parametric multimodal correlation test to explore voxel concordance-discordance in t-maps of amyloid(+) vs. amyloid(-) group differences determined for [11C]PiB (or PiB) and Florbetapir PET.
Methods: PiB and Florbetapir PET (50-70 SUVR) were performed for2 (6 NC,6 MCI) ADNI subjects, the largest exploratory sample of baseline PiB scans versus subsequent baseline Florbetapir scans (3.0±0.3 years apart). PET SUVR images were spatially normalized to a MR template using the individual’s structural MRI. Subject amyloid retention status (+/-) was based on regional iterative outlier analyses of the PiB data (9 amyloid(-),3 amyloid(+)). Voxel-wise-sample t-tests were performed to generate group difference t-maps for both PiB and Florbetapir (amyloid(+) vs. amyloid(-)). The degree of concordance-discordance was summarized using a combining t-value function within a permutation test framework [1]. Clusters of concordance-discordance were projected into normalized space and mapped to anatomical locations (pResults: Concordance was significant in high amyloid retention areas of cortex. Discordance with significant PiB (insignificant Florbetapir) arose in ventral frontal cortex, lateral temporal cortex, hippocampus and anterior cerebellum. Discordance with significant Florbetapir (insignificant PiB) arose in sensory motor cortex, occipital pole and putamen. The secondary analysis revealed a greater concordance-discordance overlap in the cortex, and discordance in the anterior-ventral striatum.
Conclusions: Group differences were highly concordant in amyloid-bearing areas of cortex. Discordance arose in areas with intra- and inter-subject variation (e.g., non-uniform retention, individual anatomical variability, low retention). Changes that may arise during the interval between PiB and Florbetapir scanning (e.g., atrophy, amyloid accumulation) may contribute to discordance.
[1] Hayasaka et al. Neuroimage (2006)


Please login to recommend the paper.


No Available Comments

Make a Comment

To make a comment you must login or register.

This paper appears in the following:


  1. News Focus: 2012 Human Amyloid Imaging Conference