Please Wait...

Effect size and spatial smoothing of ADNI Florbetapir and FDG PET

Gregory Klein, Mehul Sampat, Davis Staewen, David Scott, Joyce Suhy
BioClinica, Newark, CA, USA

We compare results of differing PET scanner spatial resolution on effect size between group differences of standard uptake value ratio (SUVR) measurements in Florbetapir (AV45) and FDG images from the ADNI database. Most reported ADNI analyses use "Level 4" preprocessed data [Koeppe 2009] smoothed to a uniform isotropic resolution of 8mm FWHM, but there is the potential that the unsmoothed data captures more sensitive information.

Freesurfer was used to obtain an ROI parcellation on T1 MRI data from 458 subjects (175 Normal, 92 EMCI, 153 LMCI, 39AD). Florbetapir and FDG data from two time points approximately 24 months apart were registered to the MRI data in T1 native space, and SUVR's were computed at each time point. Results were obtained from both ADNI level 4 smoothed data, and from unsmoothed images (ADNI level 3) [Koeppe 2009]. Cross sectional effect sizes between AD, EMCI, LMCI and Normal groups were evaluated using Cohen's d. Effect size of longitudinal change in each group was also compared. Eight reference regions (described below) were evaluated for both the Florbetapir and FDG data, Cortical ROIs for the AV45 SUVR calculation followed the [Landau 2013] method, and FDG cortical ROI selection was based on individual region effect size  (Figure 1).

SUVR Reference Regions for AV45  and  FDG Analysis
Numerous potential reference regions were evaluated:
Brainstem (BS)
Cerebellar Grey (CG)
Whole Cerebellum (WC)
Cerebellar White Matter (WMcereb)
Subcortical White Matter (WM)
Eroded Subcortical White Matter (WMeroded)
Whole Brain (WB)
Average of BS, WMeroded, WC (AvgRef)

Comparison Metric
Cohen's d effect size is the metric used for comparison of methods:

Unsmoothed data consistently showed slightly higher effect size in both the cross sectional and longitudinal analyses of the Florbetapir data, regardless of the reference region used. This trend was not seen in the FDG data, where there was little difference between the smoothed and unsmoothed data.

Cross sectional and longitudinal effect sizes in the Florbetapir analysis were largest using a reference regions including subcortical white matter for both the smoothed and unsmoothed data.

Cross sectional and longitudinal effect sizes in FDG data including subcortical white matter generally show a decreased effect size in both smoothed and unsmoothed data, indicating that these reference regions are not optimal for FDG analyses. The decreased effect size of FDG SUVRs using white matter reference regions also imply that tracer uptake in adjacent grey and white matter regions included in the SUVR calculation are positively correlated.

There appears to be a trend for increasing effect size in longitudinal and cross sectional SUVR analyses of Florbetapir PET data using unsmoothed compared to smoothed data, but not in FDG data. In a clinical trial setting, this could imply that longitudinal studies of Florbetapir can obtain greater sensitivity to detecting therapeutic effects by using PET scanners with higher resolution.

Use of unsmoothed data in the FDG analysis did not show the advantages that were seen in the Florbetapir analysis. One possible reason for this could be due to likely correlation between adjacent grey and white matter uptake in FDG images. In this case, the partial volume effect in the smoothed data would not tend to decrease any signal that might be present in differentiating AD from N patient subgroups. Conversely, for the Florbetapir data, it is the differentiation of grey from white matter uptake that distinguishes diseased from normal subjects. In this case, there is an advantage to higher spatial resolution, less smoothed data.


  1. Koeppe et al. ADNI PET Preprocessing 2009
  2. Landau et al. J Nucl Med 2013; 54:1–8.


Leader in Clinical Trial
Management Solutions

Successful clinical trials require the ability to see key details and uncover hidden insights. Bioclinica utilizes science and technology to bring clarity to clinical trials, helping companies to develop new life-improving therapies more efficiently and safely.

ASCPT 2018 Annual Meeting Join @Bioclinica Research's Open House 3/22
bioclinica (2 days ago)
Heading to DC for DIA's #Pharmacovigilance & Risk Mgmt Strategies Conf.? Visit @Bioclinica in booth 32. Preview our…
bioclinica (2 days ago)
Kevin Bishop, named a 'Top Innovator' in #clinicaltrials by @CenterWatch See how we're reducing risk to the clinic…
bioclinica (3 days ago)
RT @CenterWatch: Congratulations to all our #2018TopInovators! Read about the innovations in the #CWMonthly special edition,…
bioclinica (2 weeks ago)
RT @CenterWatch: .@Pfizer demonstrates leadership in #clinicaltrial innovation with two spots on the @CenterWatch #2018TopInnovators, https…
bioclinica (2 weeks ago)
RT @jheanne: I just contributed towards 'Joby' on Best wishes to Joby and all our @Bioclinica team members! @Oxfam…
bioclinica (2 weeks ago)

Latest Blogs:

Removing Risk from Clinical Trial Management System (CTMS) Implementations
Collaboration Between Clinical Operations and the Logistics and Supply Chain Teams is Key to Trial Success
The Value of Protocol Review
CTMS and RBM: Hot Topics at OCT Nordics in Copenhagen