Holistic neuroimaging compression to improve clinical predictivity of post-stroke mood symptoms at 6 months
Brain Structure and Function · 17.11.2026 · 231 (9)
DOI: 10.1007/s00429-026-03222-5 ↗Kurz gefasst
Fachartikel in der Zeitschrift Brain Structure and Function, online erschienen am 17.11.2026. Verfasst von Suhrit Duttagupta und 5 weiteren Autor:innen. Beteiligte Einrichtung: Univ. Bordeaux, INCIA UMR 5287, 33000, Bordeaux, France. suhrit.duttagupta@u-bordeaux.fr. Hauptthemen laut MEDLINE: Stroke, Brain, Neuroimaging, Affect. Die Arbeit nennt 48 Literaturangaben. Gefördert von: Agence Nationale de la Recherche, Ministère des Solidarités et de la Santé.
Themen: Schlaganfall
Abstract (Originalsprache)
Post-stroke mood impairments are frequent yet difficult to predict using neuroimaging markers. While voxel-level analyses have struggled to identify reliable predictors, network-level approaches may offer greater sensitivity. Accordingly, we developed a pipeline to predict post-stroke mood status using whole-brain components derived from MRI sequences routinely acquired in clinical practice. Three heterogeneous stroke cohorts (n_total = 672) were included, with mood evaluated six months post-stroke using the Hospital Anxiety and Depression-Anxiety (HAD-A), HAD-Depression (HAD-D), and Center for Epidemiologic Studies–Depression (CES-D) scales. For model training, diffusion-weighted signal from one cohort was compressed using principal component analysis (PCA) and components associated with numerical mood scores were fitted using K-Nearest Neighbour regression. Predictive performance was externally validated and compared with disconnectome-based models and models comprising baseline clinical and demographic variables. Disconnectome-based models showed minimal predictive power across datasets (R²
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Univ. Bordeaux, INCIA UMR 5287, 33000, Bordeaux, France. suhrit.duttagupta@u-bordeaux.fr.
- Groupe d'Imagerie Neurofonctionnelle, Institut des Maladies Neurodégénératives 5293, Centre National de la Recherche Scientifique, 33000, Bordeaux, France.
- Brain Connectivity and Behaviour Laboratory, Sorbonne University, 75013, Paris, France.
- Univ. Bordeaux, Neurocentre Magendie UMR 1215, 33000, Bordeaux, France.
- CHU Bordeaux, Diagnostic and Therapeutic Neuroimaging Unit, 33000, Bordeaux, France.
- Department of Neurology, Amiens University Hospital, 80000, Amiens, France.
Förderung
- Agence Nationale de la Recherche
- Ministère des Solidarités et de la Santé
Zitieren
APA
Duttagupta, S., de Schotten, M. T., Tourdias, T., Godefroy, O., Berthoz, S., & Sibon, I. (2026). Holistic neuroimaging compression to improve clinical predictivity of post-stroke mood symptoms at 6 months. Brain Structure and Function, 231(9). https://doi.org/10.1007/s00429-026-03222-5
Vancouver
Duttagupta S, de Schotten MT, Tourdias T, Godefroy O, Berthoz S, Sibon I. Holistic neuroimaging compression to improve clinical predictivity of post-stroke mood symptoms at 6 months. Brain Structure and Function. 2026;231(9). doi:10.1007/s00429-026-03222-5
BibTeX
@article{Duttagupta2026,
author = {Duttagupta, Suhrit and de Schotten, Michel Thiebaut and Tourdias, Thomas and Godefroy, Olivier and Berthoz, Sylvie and Sibon, Igor},
title = {Holistic neuroimaging compression to improve clinical predictivity of post-stroke mood symptoms at 6 months},
journal = {Brain Structure and Function},
year = {2026},
volume = {231},
number = {9},
doi = {10.1007/s00429-026-03222-5}
}
EUVIMED