Prediction of Long-term Cognitive Function After Minor Stroke Using Functional Connectivity
Renaud Lopes, Clément Bournonville, Grégory Kuchcinski, Thibaut Dondaine, Anne‐Marie Mendyk, Romain Viard, Jean‐Pierre Pruvo, Hilde Hénon, Marios K. Georgakis, Marco Duering, Martin Dichgans, Charlotte Cordonnier, X. Leclerc, Régis Bordet
Abstract
OBJECTIVE: To determine whether functional MRI connectivity can predict long-term cognitive function 36 months after minor stroke. METHODS: Seventy-two participants with first-ever stroke were included at baseline and followed up for 36 months. A ridge regression machine learning algorithm was developed and used to predict cognitive scores 36 months poststroke on the basis of the functional networks measured using MRI at 6 months (referred to here as the poststroke cognitive impairment [PSCI] network). The prediction accuracy was evaluated in 4 domains (memory, attention/executive, language, and visuospatial functions) and compared with clinical data and other functional networks. The models' statistical significance was probed with permutation tests. The potential involvement of cortical atrophy was assessed 6 months poststroke. A second, independent dataset (n = 40) was used to validate the results and assess their generalizability. RESULTS: : 0.67, 0.73, 0.55, and 0.48, respectively). The PSCI-based model was at least as accurate as models based on other functional networks or clinical data. Specific patterns were demonstrated for the 4 cognitive domains, with involvement of the left superior frontal cortex for memory, attention, and visuospatial functions. The cortical thickness 6 months poststroke was not correlated with cognitive function 36 months poststroke. The independent validation dataset gave similar results. CONCLUSIONS: A machine learning model based on the PSCI network can predict long-term cognitive outcome after stroke.