Comparison of SPM and NEUROSTAT in voxelwise statistical analysis of brain SPECT and MRI at the early stage of Alzheimer's disease

Neuroimaging plays a major role in the early diagnosis of Alzheimer's disease (AD). Recent advances in voxelwise statistical analysis after anatomic standardization of images have made this early diagnosis easier and more objective than visual inspection. We present comparative observations of NEUROSTAT, statistical parametric mapping (SPM) 99, and SPM2 in the early diagnosis of AD using brain perfusion single-photon emission computed tomography (SPECT) and magnetic resonance imaging (MRI). We performed voxel-by-voxel statistical group analysis for brain perfusion SPECT and gray matter images segmented from MRI between 61 patients with very early AD and 82 age-matched healthy volunteers. Anatomic standardization was performed using NEUROSTAT, SPM99, and SPM2 using both original and common templates. The location of significant reduction of regional cerebral blood flow (rCBF) for SPECT and gray matter concentration for MRI were identical among these three methods irrespective of the templates used. When using the original template, the significance of peak rCBF reduction in the posterior cingulate gyri was higher in SPM99 and SPM2 than that in NEUROSTAT. On the other hand, when using the common template, the significance of peak rCBF reduction in the posterior cingulate gyri was higher in NEUROSTAT and SPM2 than that in SPM99. NEUROSTAT showed almost the equal significance of peak rCBF reduction between the used templates. Almost the equal significance of reduction in gray matter concentration was observed in the parahippocampal gyri among the three methods. NEUROSTAT, SPM99, and SPM2 showed identical location of significant reductions in rCBF and gray matter concentration in very early AD patients. Used templates for anatomic standardization are relevant to the results of voxelwise statistical analysis in SPM, less prominently in SPM2 than in SPM99, whereas irrelevant in NEUROSTAT.