Dataset Information

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.

SUBMITTER: Bisenius S

PROVIDER: S-EPMC5310935 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Primary progressive aphasia is a clinical syndrome that encompasses three major phenotypes: non-fluent/agrammatic, semantic and logopenic. These clinical entities have been associated with characteristic patterns of focal grey matter atrophy in left posterior frontoinsular, anterior temporal and left temporoparietal regions, respectively. Recently, network-level dysfunction has been hypothesized but research to date has focused largely on studying grey matter damage. The aim of this study was to assess the integrity of white matter tracts in the different primary progressive aphasia subtypes. We used diffusion tensor imaging in 48 individuals: nine non-fluent, nine semantic, nine logopenic and 21 age-matched controls. Probabilistic tractography was used to identify bilateral inferior longitudinal (anterior, middle, posterior) and uncinate fasciculi (referred to as the ventral pathway); and the superior longitudinal fasciculus segmented into its frontosupramarginal, frontoangular, frontotemporal and temporoparietal components, (referred to as the dorsal pathway). We compared the tracts' mean fractional anisotropy, axial, radial and mean diffusivities for each tract in the different diagnostic categories. The most prominent white matter changes were found in the dorsal pathways in non-fluent patients, in the two ventral pathways and the temporal components of the dorsal pathways in semantic variant, and in the temporoparietal component of the dorsal bundles in logopenic patients. Each of the primary progressive aphasia variants showed different patterns of diffusion tensor metrics alterations: non-fluent patients showed the greatest changes in fractional anisotropy and radial and mean diffusivities; semantic variant patients had severe changes in all metrics; and logopenic patients had the least white matter damage, mainly involving diffusivity, with fractional anisotropy altered only in the temporoparietal component of the dorsal pathway. This study demonstrates that both careful dissection of the main language tracts and consideration of all diffusion tensor metrics are necessary to characterize the white matter changes that occur in the variants of primary progressive aphasia. These results highlight the potential value of diffusion tensor imaging as a new tool in the multimodal diagnostic evaluation of primary progressive aphasia.

Dataset Information

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

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets