EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Multivariate Searchlight Decoding of Structural Disconnection and Network Features for Predicting Post-Stroke Deficits

Multivariates Searchlight-Decoding struktureller Disconnection und Netzwerkmerkmale zur Vorhersage post-stroke Defizite

Amir Mousavi, Maedeh Khalilian, Flore Dorchies iD, Martine Roussel, Aubignat Mickael iD, Olivier Godefroy, Ardalan Aarabi iD

Journal of Neural Engineering · 11.11.2026

DOI: 10.1088/1741-2552/aec0d2 ↗

Auf einen Blick

Bei 326 Schlaganfallpatient:innen sagte ein multivariates Searchlight-Decoding motorische, exekutive und Verarbeitungs-/Aktionsgeschwindigkeits-Defizite voraus. Die motorischen Ergebnisse waren am stärksten.

Fragestellung

Kann ein multivariates Searchlight-Decoding mit struktureller Disconnection und Netzwerkmerkmalen post-stroke Defizite vorhersagen?

Methode

Analyse von 326 Schlaganfallpatient:innen der GRECogVASC-Kohorte 6 Monate nach Schlaganfall; Unterstützung durch 403 gesunde Kontrollen; SVM-Klassifikatoren mit Bootstrap-Aggregation und Mehrheitsvotum.

Ergebnisse

Für links-/rechtsseitige motorische Defizite lagen die mittleren Genauigkeiten bei 68,7–77,9 %, 66,4–74,6 % und 64,8–71,8 %. Exekutive und Verarbeitungs-/Aktionsgeschwindigkeits-Defizite lagen über Zufall, aber niedriger (54,5–66,9 %).

Einschränkungen

Die kognitiven Ergebnisse erfordern unabhängige Replikation vor neuroanatomischer oder klinischer Interpretation.

Schlussfolgerung der Autor:innen

Die integrierte Struktur-Disconnection-/Netzwerk-Analyse bietet einen inversen Inferenzansatz als Ergänzung zu Läsions- und Disconnection-Symptom-Mapping; die Motorbefunde stützen das Konzept.

Kernaussagen

  • Motorische Defizite wurden mit hoherer Genauigkeit vorhergesagt als kognitive Defizite.
  • Predictive Regionen lagen v. a. im kontralateralen Gyrus praecentralis, der Kapsel interna und dem Tractus corticospinalis.
  • Die Methode reduzierte die Voxel-Dimensionalität um bis zu 95,4 %.

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Neural Engineering, online erschienen am 11.11.2026. Verfasst von Amir Mousavi und 6 weiteren Autor:innen. Beteiligte Einrichtung: Laboratory of Functional Neuroscience and Pathologies (UR UPJV 4559), University of Picardie-Jules Verne, University Research Center (CURS) CHU AMIEN…. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: French Ministry of Health.

Auf einen BlickAbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objective. Lesion-symptom mapping and structural disconnection analyses have advanced our understanding of post-stroke deficits but remain predominantly based on forward inference, limiting identification of regions specifically predictive of behavioral impairment. We developed a multivariate voxelwise searchlight decoding framework for inverse inference integrating probabilistic structural disconnection with network-topological features.
Approach. The analysis included 326 stroke patients from the GRECogVASC cohort assessed six months after stroke. For each patient, an age-specific probabilistic structural disconnection map was reconstructed using normative diffusion MRI tractography from 403 healthy controls. Within each voxel-centered searchlight, four features were extracted: disconnection probability, degree, clustering coefficient, and closeness centrality. Support vector machine classifiers were trained within deficit-specific decoding spaces using balanced bootstrap aggregation and majority voting to predict motor, executive, and processing-action-speed impairments. Statistical significance was assessed using 1,000 permutations, retaining voxels that consistently performed above chance across bootstrap iterations. Decoding-derived regions were subsequently assessed for within-cohort individual-level classification. 
Main results. Initial validation focused on left and right motor impairments, identifying predictive regions primarily within the contralateral precentral gyrus, internal capsule, and corticospinal tract, with mean accuracy ranges of 68.7–77.9%, 66.4–74.6%, and 64.8–71.8%, respectively. At the individual level, decoding-derived masks preserved overall classification performance, with accuracy improvements of up to 2.6% for individual deficits while reducing voxel dimensionality by up to 95.4%. Executive dysfunction and processing-action-speed deficits showed statistically above-chance but lower accuracies (54.5–66.9%). Executive dysfunction involved predominantly left-lateralized frontotemporal regions, whereas processing-action-speed deficits involved distributed fronto-insular, temporal, parietal, visual-association, and commissural pathways.
Significance. The multivariate searchlight decoding framework integrating structural disconnection and network topology provides an inverse-inference approach complementary to lesion- and disconnection-symptom mapping for identifying spatially specific predictive patterns after stroke. The strong motor findings provide proof-of-concept support for the approach, whereas the cognitive results support its extension to distributed behavioral domains while underscoring the need for independent replication before neuroanatomical or clinical interpretation (Clinical trial registration: NCT01339195).

Schlagwörter

Stichwörter der Autor:innen

Strokebrain networksexecutive dysfunctiongraph metricsmotor impairmentsearchlight decodingstructural disconnection

Autor:innen und Institutionen

  • Laboratory of Functional Neuroscience and Pathologies (UR UPJV 4559), University of Picardie-Jules Verne, University Research Center (CURS) CHU AMIENS - SITE SUD Avenue René Laennec, Amiens, 80054, France.
  • Laboratory of Functional Neuroscience and Pathologies (UR UPJV 4559), University of Picardie-Jules Verne, University Research Center (CURS) CHU AMIENS - SITE SUD Avenue René Laennec, Amiens, Hauts-de-France, 80054, Fran…
  • University of Picardie-Jules Verne, University Research Center (CURS) CHU AMIENS - SITE SUD Avenue René Laennec, Amiens, Hauts-de-France, 80054, France.
  • University of Picardy Jules Verne, University Research Center (CURS) CHU AMIENS - SITE SUD Avenue René Laennec, Amiens, Hauts-de-France, 80054, France.
  • Laboratory of Functional Neuroscience and Pathologies (UR UPJV 4559), University Research Center (CURS), University of Picardie-Jules Verne, University Research Center (CURS) CHU AMIENS - SITE SUD Avenue René Laennec, A…

ORCID-Profile: Flore Dorchies, Aubignat Mickael, Ardalan Aarabi

Förderung

  • French Ministry of Health

Zitieren

APA

Mousavi, A., Khalilian, M., Dorchies, F., Roussel, M., Mickael, A., Godefroy, O., & Aarabi, A. (2026). Multivariate Searchlight Decoding of Structural Disconnection and Network Features for Predicting Post-Stroke Deficits. Journal of Neural Engineering. https://doi.org/10.1088/1741-2552/aec0d2

Vancouver

Mousavi A, Khalilian M, Dorchies F, Roussel M, Mickael A, Godefroy O, et al. Multivariate Searchlight Decoding of Structural Disconnection and Network Features for Predicting Post-Stroke Deficits. Journal of Neural Engineering. 2026. doi:10.1088/1741-2552/aec0d2

BibTeX
@article{Mousavi2026,
  author = {Mousavi, Amir and Khalilian, Maedeh and Dorchies, Flore and Roussel, Martine and Mickael, Aubignat and Godefroy, Olivier and Aarabi, Ardalan},
  title = {Multivariate Searchlight Decoding of Structural Disconnection and Network Features for Predicting Post-Stroke Deficits},
  journal = {Journal of Neural Engineering},
  year = {2026},
  doi = {10.1088/1741-2552/aec0d2}
}
Kostenlos starten →