Predicting primary progressive aphasias with support vector machine approaches in structural MRI data

Abstract

Primary progressive aphasia (PPA) encompasses the three subtypes nonfluent/agrammatic variant PPA, semantic variant PPA, and the logopenic variant PPA, which are characterized by distinct patterns of language difficulties and regional brain atrophy. To validate the potential of structural magnetic resonance imaging data for early individual diagnosis, we used support vector machine classification on grey matter density maps obtained by voxel-based morphometry analysis to discriminate PPA subtypes (44 patients: 16 nonfluent/agrammatic variant PPA, 17 semantic variant PPA, 11 logopenic variant PPA) from 20 healthy controls (matched for sample size, age, and gender) in the cohort of the multi-center study of the German consortium for frontotemporal lobar degeneration. Here, we compared a whole-brain with a meta-analysis-based disease-specific regions-of-interest approach for support vector machine classification. We also used support vector machine classification to discriminate the three PPA subtypes from each other. Whole brain support vector machine classification enabled a very high accuracy between 91 and 97% for identifying specific PPA subtypes vs. healthy controls, and 78/95% for the discrimination between semantic variant vs. nonfluent/agrammatic or logopenic PPA variants. Only for the discrimination between nonfluent/agrammatic and logopenic PPA variants accuracy was low with 55%. Interestingly, the regions that contributed the most to the support vector machine classification of patients corresponded largely to the regions that were atrophic in these patients as revealed by group comparisons. Although the whole brain approach took also into account regions that were not covered in the regions-of-interest approach, both approaches showed similar accuracies due to the disease-specificity of the selected networks. Conclusion, support vector machine classification of multi-center structural magnetic resonance imaging data enables prediction of PPA subtypes with a very high accuracy paving the road for its application in clinical settings.

Graded, multi-dimensional intragroup and intergroup variations in primary progressive aphasia and post-stroke aphasia

“…In fact, the concept of 'SD' seems to be a uniquely useful pointer for the exclusive region of the multi-dimensional space occupied by these cases. This aligns with (i) the original descriptions of SD, in particular the selective nature of their semantic impairment (Hodges et al, 1992;Snowden et al, 1989;Warrington, 1975), and (ii) previous work showing that SD is distinct from other forms of PPA; Bisenius et al (2017) found that SD was the most readily differentiable subtype of PPA using Support Vector Machine approaches to evaluate the consensus criteria for PPA. Hoffman et al (2017) applied k-means clustering to behavioural data in PPA and found that of their three-cluster solution, only one cluster was selective for a particular subtype of PPA and this was the SD cluster.…”

Morphometric MRI as a diagnostic biomarker of frontotemporal dementia: A systematic review to determine clinical applicability

“…Four studies included classifications of PPA ( Bisenius et al, 2017 ; Chow et al, 2008 ; Tahmasian et al, 2016 ; Wilson et al, 2009 ) ( Table 5 ). Two studies classified each PPA subtype against controls using SVM of GM atrophy, with moderate to high accuracy across studies (accuracy ranged from 84 to 100%) ( Bisenius et al, 2017 ; Wilson et al, 2009 ). Both studies also classified subtypes against each other, with varying results.…”

Graded, multidimensional intra- and intergroup variations in primary progressive aphasia and post-stroke aphasia

“…This aligns with (i) the original descriptions of semantic dementia, in particular the selective nature of their semantic impairment ( Warrington, 1975 ; Snowden et al , 1989 ; Hodges et al , 1992 ); and (ii) previous work showing that semantic dementia is distinct from other forms of PPA. Bisenius et al (2017) found that semantic dementia was the most readily differentiable subtype of PPA using support vector machine approaches to evaluate the consensus criteria for PPA. Hoffman et al (2017) applied k-means clustering to behavioural data in PPA and found that of their three-cluster solution, only one cluster was selective for a particular subtype of PPA and this was the semantic dementia cases.…”