LDA PMOD-R Implementation

<< Click to Display Table of Contents >>

Navigation:  R Statistics in PMOD > Analysis Scripts for Aggregates > Linear Models: Statistical Analysis of Regional Imaging Data > Discrimination Analysis >

LDA PMOD-R Implementation

The example data were downloaded from the ADNI website (http://adni.loni.usc.edu) in January 2016.

ADNI acknowledgement

Data used in preparation of this section were obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database (adni.loni.usc.edu). The investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. A complete listing of ADNI investigators can be found at:


The PET scans represented the brain activity 30-60 minutes after 18F-FDG injection and were downloaded as pre-processed data (standardized image and voxel size). Briefly, this data is already  scaled to a common global averaged value, reoriented into a standard 160 x 160 x96 voxel imaged grid (voxel size, 1.5 x 1.5 x 1.5 mm) along the anterior commissure-posterior commissure (AC-PC) plane and formatted as DICOM or ECAT files.

The downloaded PET images were then processed using the PNEURO module in the PMOD software package (version 3.9; PMOD technologies). Particularly, the AAL-Atlas VOI was adjusted to the individual data with an SPM 5 equivalent normalization procedure and average regional statistics were calculated in each VOI.

The baseline FDG data in two patient groups with mild cognitive impairment (MCI) are used for the present analysis: one group of 129 subjects with stable MCI (age 74.93 ± 7.45 years), the other of 78 subjects (age 75.37 ± 6.56 years) progressing to dementia within 2 years of clinical follow-up. We are focusing on two brain areas known to be predictors of progression, the posterior cingulate (PCC) and the angular cortex (AngC). The respective variables are Cingulum_Post_l_C35 and Cingulum_Post_r_C36 (left and right PCC) and Angular_l_C65 and Angular_r_C66 (left and right AngC). Values were normalized to pons as reference region.