Dataset Information

Enhancing the Classification of Aphasia: A Statistical Analysis Using Connected Speech.

ABSTRACT:

Background

Large shared databases and automated language analyses allow for the application of new data analysis techniques that can shed new light on the connected speech of people with aphasia (PWA).

Aims

To identify coherent clusters of PWA based on language output using unsupervised statistical algorithms and to identify features that are most strongly associated with those clusters.

Methods & procedures

Clustering and classification methods were applied to language production data from 168 PWA. Language samples were from a standard discourse protocol tapping four genres: free speech personal narratives, picture descriptions, Cinderella storytelling, procedural discourse.

Outcomes & results

Seven distinct clusters of PWA were identified by the K-means algorithm. Using the random forests algorithm, a classification tree was proposed and validated, showing 91% agreement with the cluster assignments. This representative tree used only two variables to divide the data into distinct groups: total words from free speech tasks and total closed class words from the Cinderella storytelling task.

Conclusion

Connected speech data can be used to distinguish PWA into coherent groups, providing insight into traditional aphasia classifications, factors that may guide discourse research and clinical work.

SUBMITTER: Fromm D

PROVIDER: S-EPMC9708051 | biostudies-literature | 2022

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Enhancing the Classification of Aphasia: A Statistical Analysis Using Connected Speech.

Fromm Davida D Greenhouse Joel J Pudil Mitchell M Shi Yichun Y MacWhinney Brian B

Aphasiology 20210921 12

<h4>Background</h4>Large shared databases and automated language analyses allow for the application of new data analysis techniques that can shed new light on the connected speech of people with aphasia (PWA).<h4>Aims</h4>To identify coherent clusters of PWA based on language output using unsupervised statistical algorithms and to identify features that are most strongly associated with those clusters.<h4>Methods & procedures</h4>Clustering and classification methods were applied to language pro ...[more]

PMID: 36457942

Similar Datasets

Project description:Primary progressive aphasia is a clinical syndrome defined by progressive deficits isolated to speech and/or language, and can be classified into non-fluent, semantic and logopenic variants based on motor speech, linguistic and cognitive features. The connected speech of patients with primary progressive aphasia has often been dichotomized simply as 'fluent' or 'non-fluent', however fluency is a multidimensional construct that encompasses features such as speech rate, phrase length, articulatory agility and syntactic structure, which are not always impacted in parallel. In this study, our first objective was to improve the characterization of connected speech production in each variant of primary progressive aphasia, by quantifying speech output along a number of motor speech and linguistic dimensions simultaneously. Secondly, we aimed to determine the neuroanatomical correlates of changes along these different dimensions. We recorded, transcribed and analysed speech samples for 50 patients with primary progressive aphasia, along with neurodegenerative and normal control groups. Patients were scanned with magnetic resonance imaging, and voxel-based morphometry was used to identify regions where atrophy correlated significantly with motor speech and linguistic features. Speech samples in patients with the non-fluent variant were characterized by slow rate, distortions, syntactic errors and reduced complexity. In contrast, patients with the semantic variant exhibited normal rate and very few speech or syntactic errors, but showed increased proportions of closed class words, pronouns and verbs, and higher frequency nouns, reflecting lexical retrieval deficits. In patients with the logopenic variant, speech rate (a common proxy for fluency) was intermediate between the other two variants, but distortions and syntactic errors were less common than in the non-fluent variant, while lexical access was less impaired than in the semantic variant. Reduced speech rate was linked with atrophy to a wide range of both anterior and posterior language regions, but specific deficits had more circumscribed anatomical correlates. Frontal regions were associated with motor speech and syntactic processes, anterior and inferior temporal regions with lexical retrieval, and posterior temporal regions with phonological errors and several other types of disruptions to fluency. These findings demonstrate that a multidimensional quantification of connected speech production is necessary to characterize the differences between the speech patterns of each primary progressive aphasic variant adequately, and to reveal associations between particular aspects of connected speech and specific components of the neural network for speech production.

Project description:The consensus criteria for the diagnosis and classification of primary progressive aphasia (PPA) have served as an important tool in studying this group of disorders. However, a large proportion of patients remain unclassifiable whilst others simultaneously meet criteria for multiple subtypes. We prospectively evaluated a large cohort of patients with degenerative aphasia and/or apraxia of speech using multidisciplinary clinical assessments and multimodal imaging. Blinded diagnoses were made using operational definitions with important differences compared to the consensus criteria. Of the 130 included patients, 40 were diagnosed with progressive apraxia of speech (PAOS), 12 with progressive agrammatic aphasia, 9 with semantic dementia, 52 with logopenic progressive aphasia, and 4 with progressive fluent aphasia, while 13 were unclassified. The PAOS and progressive fluent aphasia groups were least impaired. Performance on repetition and sentence comprehension was especially poor in the logopenic group. The semantic and progressive fluent aphasia groups had prominent anomia, but only semantic subjects had loss of word meaning and object knowledge. Distinct patterns of grey matter loss and white matter changes were found in all groups compared to controls. PAOS subjects had bilateral frontal grey matter loss, including the premotor and supplementary motor areas, and bilateral frontal white matter involvement. The agrammatic group had more widespread, predominantly left sided grey matter loss and white matter abnormalities. Semantic subjects had bitemporal grey matter loss and white matter changes, including the uncinate and inferior occipitofrontal fasciculi, whereas progressive fluent subjects only had left sided temporal involvement. Logopenic subjects had diffuse and bilateral grey matter loss and diffusion tensor abnormalities, maximal in the posterior temporal region. A diagnosis of logopenic aphasia was strongly associated with being amyloid positive (46/52 positive). Our findings support consideration of an alternative way of identifying and categorizing subtypes of degenerative speech and language disorders.

Project description:Speech segmentation is one of the initial and mandatory phases of language learning. Although some people with aphasia have shown a preserved ability to learn novel words, their speech segmentation abilities have not been explored.We examined the ability of individuals with chronic aphasia to segment words from running speech via statistical learning. We also explored the relationships between speech segmentation and aphasia severity, and short-term memory capacity. We further examined the role of lesion location in speech segmentation and short-term memory performance.The experimental task was first validated with a group of young adults (n = 120). Participants with chronic aphasia (n = 14) were exposed to an artificial language and were evaluated in their ability to segment words using a speech segmentation test. Their performance was contrasted against chance level and compared to that of a group of elderly matched controls (n = 14) using group and case-by-case analyses.As a group, participants with aphasia were significantly above chance level in their ability to segment words from the novel language and did not significantly differ from the group of elderly controls. Speech segmentation ability in the aphasic participants was not associated with aphasia severity although it significantly correlated with word pointing span, a measure of verbal short-term memory. Case-by-case analyses identified four individuals with aphasia who performed above chance level on the speech segmentation task, all with predominantly posterior lesions and mild fluent aphasia. Their short-term memory capacity was also better preserved than in the rest of the group.Our findings indicate that speech segmentation via statistical learning can remain functional in people with chronic aphasia and suggest that this initial language learning mechanism is associated with the functionality of the verbal short-term memory system and the integrity of the left inferior frontal region.

Dataset Information

Enhancing the Classification of Aphasia: A Statistical Analysis Using Connected Speech.

Background

Aims

Methods & procedures

Outcomes & results

Conclusion

Publications

Enhancing the Classification of Aphasia: A Statistical Analysis Using Connected Speech.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets