CoChat
Sign Up
Cited by 132Neurobiology of Language and BilingualismDementia and Cognitive Impairment ResearchReading and Literacy Development

Predicting aphasia type from brain damage measured with structural MRI

Open access
Yourganov, Grigori, Smith, Kimberly G., Fridriksson, Julius, Rorden, Chris·2015·Cortex
DOI 10.1016/j.cortex.2015.09.005PMC PMC4689665
Open this paper in CoChat — free account

Abstract

Chronic aphasia is a common consequence of a left-hemisphere stroke. Since the early insights by Broca and Wernicke, studying the relationship between the loci of cortical damage and patterns of language impairment has been one of the concerns of aphasiology. We utilized multivariate classification in a cross-validation framework to predict the type of chronic aphasia from the spatial pattern of brain damage. Our sample consisted of 98 patients with five types of aphasia (Broca's, Wernicke's, global, conduction, and anomic), classified based on scores on the Western Aphasia Battery (WAB). Binary lesion maps were obtained from structural MRI scans (obtained at least 6 months poststroke, and within 2 days of behavioural assessment); after spatial normalization, the lesions were parcellated into a disjoint set of brain areas. The proportion of damage to the brain areas was used to classify patients' aphasia type. To create this parcellation, we relied on five brain atlases; our classifier (support vector machine - SVM) could differentiate between different kinds of aphasia using any of the five parcellations. In our sample, the best classification accuracy was obtained when using a novel parcellation that combined two previously published brain atlases, with the first atlas providing the segmentation of grey matter, and the second atlas used to segment the white matter. For each aphasia type, we computed the relative importance of different brain areas for distinguishing it from other aphasia types; our findings were consistent with previously published reports of lesion locations implicated in different types of aphasia. Overall, our results revealed that automated multivariate classification could distinguish between aphasia types based on damage to atlas-defined brain areas.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1016/j.cortex.2015.09.005

Cites

71 of 71 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Four included studies constructed diagnostic and classification models of aphasia by evaluating general demographic data, neuroimaging data, and clinical indicators of patients.

    “Four studies (19, 22, 30, 37) constructed diagnostic and classification models of aphasia by evaluating general demographic data, neuroimaging data, and clinical indicators of patients, but there were differences in model function, fitting, and application effect.”

    From Machine learning models in post-stroke aphasia: a scoping review (2026) · cites this paper as [37]

    The Methods (sections 2.1 and 2.4; page unavailable) describe an aphasia-type classifier using MRI-derived brain damage, with types established by Western Aphasia Battery assessment. Demographic data are reported for participants, but section 2.4 specifies brain damage—not demographic or clinical indicators—as the classifier input. This source also cannot establish what the other three cited studies did. (LLM, conf=0.88)

    Experimental — accuracy not yet measured

  • The model constructed by Yourganov et al. can automatically identify classical types such as Broca's aphasia and Wernicke's aphasia according to the proportion of brain damage.

    “For example, the model constructed by Yourganov et al. (37) can automatically identify classical types such as Broca’s aphasia and Wernicke’s aphasia according to the proportion of brain damage.”

    From Machine learning models in post-stroke aphasia: a scoping review (2026) · cites this paper as [37]

  • Three models focus on subtype classification or diagnosis of aphasia according to brain damage patterns or clinical indicators.

    “In terms of model function, three models (19, 30, 37) focus on subtype classification or diagnosis of aphasia according to brain damage patterns or clinical indicators.”

    From Machine learning models in post-stroke aphasia: a scoping review (2026) · cites this paper as [37]

  • In diagnosis and classification, most existing models follow the traditional clinical classification (19, 37).

    “In diagnosis and classification, most of the existing models follow the traditional clinical classification (19, 37)”

    From Machine learning models in post-stroke aphasia: a scoping review (2026) · cites this paper as [37]

  • Yourganov et al. reported an accuracy rate of 93.56%.

    “Yourganov et al. (37) reported an accuracy rate of 93.56%”

    From Machine learning models in post-stroke aphasia: a scoping review (2026) · cites this paper as [37]

    The Results section describes accuracy by pair of aphasia types and brain atlas, but the passages retrieved from our copy do not report 93.56%. Searches for that figure and for accuracy results did not locate it, so the specific rate cannot be verified from our copy. (LLM, conf=0.70)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

2 of the 73 references couldn’t be matched to a known paper.

Research works printed but not linked: 3

  1. Davis GA 2007 Aphasiology: Disorders and clinical practice Pearson College Division
  2. Kertesz A 2006 Western Aphasia Battery–Revised (WAB-R) Pearson San Antonio
  3. Reineck LA Agarwal S Hillis AE 2005 “Diffusion-clinical mismatch” is associated with potential for early recovery of aphasia Neurology 64 5 828 833 15753418

Cited by

Sorted by citations

Related

Authors
Yourganov, G., Smith, K. G., Fridriksson, J., & Rorden, C.
Affiliations
University of South Carolina (US)
Corresponding author
Yourganov, Grigori
ORCID
Venue
Cortex
Publisher
Elsevier BV
Published
2015-12
Volume
73
Pages
203-215
ISSN
0010-9452 (ISSN-L)
DOI
10.1016/j.cortex.2015.09.005
PMC
PMC4689665
OpenAlex
W1785208788
Cited by
132
References
73
Citations by year
2026: 8 · 2025: 12 · 2024: 13 · 2023: 11 · 2022: 9 · 2021: 14 · 2020: 18 · 2019: 17 · 2018: 14 · 2017: 12 · 2016: 4
FWCI
6.01
Type
journal article
Language
en
Open access
Licence
www.elsevier.com/tdm/userlicense/1.0/ (tdm)
Locations
  • Cortex (publishedVersion)
  • PubMed (publishedVersion)
  • PubMed Central (acceptedVersion)
Funders
National Institutes of Health (R01 DC009571)
Topics
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Neurobiology of Language and Bilingualism
  • Health Sciences › Medicine › Psychiatry and Mental health › Dementia and Cognitive Impairment Research
  • Social Sciences › Psychology › Developmental and Educational Psychology › Reading and Literacy Development
Keywords
chronic aphasia, left hemisphere stroke, lesion localization, structural MRI, brain atlas, brain parcellation, gray matter segmentation, white matter segmentation, support vector machine, Broca's aphasia, Wernicke's aphasia, global aphasia, conduction aphasia, anomic aphasia, Western Aphasia Battery
MeSH terms
Adult, Aged, Aged, 80 and over, Aphasia, Broca, Aphasia, Wernicke, Brain, Brain Injuries, Diagnosis, Differential, Female, Humans, Magnetic Resonance Imaging, Male, Middle Aged, Stroke

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review