Using in vivo functional and structural connectivity to predict chronic stroke aphasia deficits

Abstract

Focal brain damage caused by stroke can result in aphasia and advances in cognitive neuroscience suggest that impairment may be associated with network-level disorder rather than just circumscribed cortical damage. A number of studies have shown meaningful relationships between brain-behaviour using lesions; however only a handful of studies have incorporated in-vivo structural and functional connectivity. Patients with chronic post-stroke aphasia were assessed with structural (N = 68) and functional (N = 39) MRI to assess whether predicting performance can be improved with multiple modalities and if additional variance can be explained compared to lesion models alone. These neural measurements were used to construct models to predict four key language-cognitive factors: 1) phonology, 2) semantics, 3) executive function, and 4) fluency. Our results showed that each factor (except executive ability) could be significantly related to each neural measurement alone; however, structural and functional connectivity models did not explain additional variance above the lesion models. We did find evidence that the structural and functional predictors may be linked to the core lesion sites. First, the predictive functional connectivity features were found to be located within functional resting state networks identified in healthy controls, suggesting that the result might reflect functionally-specific reorganisation (damage to a node within a network can result in disruption to the entire network). Second, predictive structural connectivity features were located within core lesion sites, suggesting that multi-modal information may be redundant in prediction modelling. In addition, we observed that the optimum sparsity within the regularised regression models differed for each behavioural component and across different imaging features, suggesting that future studies should consider optimising hyperparameters related to sparsity per target. Together, the results indicate that the observed network-level disruption was predicted by the lesion alone and does not significantly improve model performance in predicting the profile of language impairment.

Predicting individual long-term prognosis of spatial neglect based on acute stroke patient data

“…These include, for example, premorbid brain atrophy (Levine et al, 1986), burden of white matter hyperintensities (Kamakura et al, 2017), fractional anisotropy values (Lunven et al, 2015), anosognosia (Stone et al, 1992), acute visual field defects (Samuelsson et al, 1997), acute allocentric (but not egocentric) neglect severity (Moore et al, 2021), as well as activation patterns and functional connectivity (Cao et al, 2022; Umarova et al, 2016). Changes in structural connectivity might also play a role (see Talozzi et al, 2023), although, for post-stroke aphasia, this measure could not add predictive value to models using the lesion information itself (Halai et al, 2020; Zhao et al, 2023). Future studies should investigate whether (some of) these or other potentially prognostic factors can add predictive information to the best-performing model(s) of the current study.…”

Structural disconnections caused by white matter hyperintensities in post-stroke spatial neglect

“…Feature selection may be a promising option to identify the most predictive variable combination. However, some research already indicated that lesion-induced disconnectivity might be unable to give further information in addition to lesion anatomy, since 26 the former is dependent on the latter (Halai et al, 2020;Hope et al, 2018;Zhao et al, 2023). On the other hand, a very recent investigation revealed that the combination of structural and functional disconnectivity (and demographics) achieved more accurate predictions than models with either one of them (and demographics) (Petersen et al, 2024).…”

Distance-related functional reorganization predicts motor outcome in stroke patients

“…During the non-acute stroke stage, survivors continue to undergo a dynamic process of functional reorganization associated with post-stroke recovery [ 1 , 2 ]. Resting-state fMRI has emerged as a promising avenue to explore brain functional integration and separation after stroke [ 3 ]. Our prior resting-state fMRI studies found decreased functional connectivity (FC) between hippocampal subfields and left postcentral gyrus as well as right middle occipital gyrus [ 4 ], between cerebellum posterior lobe and left precentral gyrus, inferior frontal gyrus as well as middle temporal gyrus [ 5 ], and increased FC from the ipsilesional primary motor cortex to the ipsilesional occipital lobes [ 6 ] in non-acute stroke patients compared with healthy controls (HCs).…”