Covariance 99mTc-exametazime SPECT patterns in Alzheimer's disease and dementia with Lewy bodies: utility in differential diagnosis.

(99m)Tc-exametazime single photon emission computed tomography (SPECT) scans of 36 patients with Alzheimer's disease (AD) and 30 with dementia with Lewy bodies (DLB) underwent region of interest (ROI) and principal component analysis (PCA). Principal component analysis was performed on the entire ROI data set. Principal components (PCs) were obtained, representing common intercorrelated regions in AD and DLB. Topographic expression that signified the extent to which a participant expressed the topographic covariance pattern was derived and used as a discriminatory variable. Principal components were identified, accounting for 77% of total data variance. Significant (PC x group) interaction was observed (P < .001). Topographic expression was significantly higher in DLB than AD (F(1,64) = 21.6, P < .001), and differentiated DLB from AD with sensitivity 73% specificity 72%. Calculating the topographic expression in an independent data set of 48 patients with AD and 23 with DLB gave sensitivity = 70%, specificity = 67%. Principal component analysis captures additional sources of variance and if perfusion SPECT is the only scan available, this procedure may offer extra information. [ABSTRACT FROM AUTHOR]