Premium
P4‐311: EVALUATION OF CANNABINOID RECEPTOR 2 PET IMAGING TRACERS IN ALZHEIMER'S DISEASE BRAIN
Author(s) -
Ni Ruiqing,
Haider Ahmed,
Herde Adrianne,
Klohs Jan,
Mu Linjing,
Ametamey Simon Mensah
Publication year - 2018
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2018.07.134
Subject(s) - hippocampus , neuroinflammation , medicine , endocrinology , in vivo , microglia , pathology , temporal cortex , biology , neuroscience , inflammation , microbiology and biotechnology
minute sub-frameof the 20-minutePETacquisitionwas selectedas the input low-dose image (Figure 2). The synthesized PET image and the low-dose ADNI PET images within a FreeSurfer-derived brain mask were compared to the original 20-minute “full-dose” ADNI image using themetrics peak signal-to-noise ratio (PSNR), structural similarity (SSIM), and root mean square error (RMSE). To assess image quality of the synthesized imagescompared to their low-dose counterparts, the relative change (RC) of mean graymatter (GM) uptake and the SD of white matter (WM) uptake, using the full-dose image as a standard, as well as SNR (ratio of mean GM toWMSD), were calculated.Results: The image quality metrics showed lower PSNR and greater RMSE but higher structural similarity for the synthesized images (Figure 3). The regional metrics also showed that the synthesized images were able to achieve higher SNR (Figure 4). Conclusions:We successfully applied the pre-trained network to ADNI datasets, which used a different F tracer and had a different dose reduction factor. The synthesized images showed comparable tracer activity with the full-dose image and higher structural similarity while achieving noise reduction compared to the low-dose image. P4-311 EVALUATION OF CANNABINOID RECEPTOR 2 PET IMAGING TRACERS IN ALZHEIMER’S DISEASE BRAIN Ruiqing Ni, Ahmed Haider, Adrianne Herde, Jan Klohs, Linjing Mu, Simon Mensah Ametamey, ETH Zurich & University of Zurich, Zurich, Switzerland; ETH Zurich, Zurich, Switzerland. Contact e-mail: ni@biomed.ee.ethz.ch