Skip to main content
. 2022 Aug 2;99(5):e500–e511. doi: 10.1212/WNL.0000000000200750

Figure 3. Binary Logistic Regression Results.

Figure 3

(A) Classification between patients with nfvPPA and HCs was significant for the automated measure of the articulation rate (top inset) as well as the perceptual measures of apraxia of speech severity (middle inset) and dysarthria severity (bottom inset), with maximal AUC score for the former. (B) Classification between patients with nfvPPA-CBD and nfvPPA-PSP was significant for the automated measure of the articulation rate (top inset) but not for the perceptual measures of apraxia of speech severity (middle inset) and dysarthria severity (bottom inset), the former yielding a robust AUC value. AUC = area under the ROC curve; HCs = healthy controls; nfvPPA = nonfluent variant primary progressive aphasia; nfvPPA-CBD = nonfluent variant primary progressive aphasia with corticobasal degeneration pathology; nfvPPA-PSP = nonfluent variant primary progressive aphasia with progressive supranuclear palsy pathology; ROC = receiving operating characteristic; svPPA = semantic variant primary progressive aphasia.