Hierarchy of connectivity-function relationship of the human cortex revealed through predicting activity across functional domains

Wu, Dongya; Fan, Lingzhong; Song, Ming; Wang, Haiyan; Chu, Congying; Yu, Shan; Jiang, Tianzi

doi:10.1101/591776

Cited by 8 publications

(12 citation statements)

References 60 publications

(72 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…35 In addition, it is expected that white matter disruption induces functional changes at the cortical level. 36 Therefore, we performed a complementary analysis aimed to integrate functional and structural maps. We defined significant functional regions and adjacent white matter voxels as seeds for tractography in the seed tracking algorithm provided by DSI Studio (http://dsi-studio.labsolver.…”

Section: Functional Connectivity Analysismentioning

confidence: 99%

Unravelling the Neural Basis of Spatial Delusions After Stroke

Alves

Fonseca

Silva

et al. 2021

Annals of Neurology

View full text Add to dashboard Cite

Objective Knowing explicitly where we are is an interpretation of our spatial representations. Reduplicative paramnesia is a disrupting syndrome in which patients present a firm belief of spatial mislocation. Here, we studied the largest sample of patients with delusional misidentifications of space (ie, reduplicative paramnesia) after stroke to shed light on their neurobiology. Methods In a prospective, cumulative, case‐control study, we screened 400 patients with acute right‐hemispheric stroke. We included 64 cases and 233 controls. First, lesions were delimited and normalized. Then, we computed structural and functional disconnection maps using methods of lesion‐track and network‐mapping. The maps were compared, controlling for confounders. Second, we built a multivariate logistic model, including clinical, behavioral, and neuroimaging data. Finally, we performed a nested cross‐validation of the model with a support‐vector machine analysis. Results The most frequent misidentification subtype was confabulatory mislocation (56%), followed by place reduplication (19%), and chimeric assimilation (13%). Our results indicate that structural disconnection is the strongest predictor of the syndrome and included 2 distinct streams, connecting right fronto‐thalamic and right occipitotemporal structures. In the multivariate model, the independent predictors of reduplicative paramnesia were the structural disconnection map, lesion sparing of right dorsal fronto‐parietal regions, age, and anosognosia. Good discrimination accuracy was demonstrated (area under the curve = 0.80 [0.75–0.85]). Interpretation Our results localize the anatomic circuits that may have a role in the abnormal spatial‐emotional binding and in the defective updating of spatial representations underlying reduplicative paramnesia. This novel data may contribute to better understand the pathophysiology of delusional syndromes after stroke. ANN NEUROL 2021;89:1181–1194

show abstract

Section: Functional Connectivity Analysismentioning

confidence: 99%

Unravelling the Neural Basis of Spatial Delusions After Stroke

Alves

Fonseca

Silva

et al. 2021

Annals of Neurology

View full text Add to dashboard Cite

show abstract

“…Among these connectivity-based methods, functionalfingerprint-based methods are dominant and direct in the cerebral cartography, providing a lot of perspective and practical works for the clinics [13,37,40]. On the other side, structure covariations and anatomical fingerprints could reflect the individual functional variability to some extent [29,42,66], but are still less investigated and involved in the brain atlas individualization. The anatomical fingerprints have the advantages in reproducibility in many studies, compared with functional fingerprints [66].…”

Section: Discussionmentioning

confidence: 99%

“…Many studies showed there exists a structure-function relationship and the identification of functional area through anatomical fingerprint [28,29,47,48]. In this study, the fibertract-based connectivity fingerprint was calculated, as the one of input in the graph convolutional network (shown in Fig.…”

Section: B Connectivity Fingerprints Of Fiber-tract Embeddingmentioning

confidence: 99%

“…First, the structure-function relationship has been widely investigated in neuroscience research for decades. A series of studies showed that each functional area has a unique anatomical connectivity fingerprint [29,41] and that the brain activations of a functional area can be inferred from its connectivity fingerprint [28,42]. A precise anatomical and biological ROIs (e.g., fiber tracts), rather than template-based locations for the anatomical fingerprints could reflect a reliable and explainable variations for each group area.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

BAI-Net: Individualized Anatomical Cerebral Cartography using Graph Neural Network

Zhang

Cheng

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Individualized human cerebral cartography has drawn considerable attention in neuroscience research. But many challenges remain, mainly due to large cross-subject variations in brain morphology, connectivity, and function. We developed a new tool called brain atlas individualization network (BAI-Net) that automatically parcels individual cerebral cortex into segregated functional areas using structural and diffusion MRIs. The presented method integrates the group priors into the loss function of graph convolution network, and learns the connectivity context of anatomical fingerprints for each area on individual brain graphs. The presented model provides reliable, efficient and explainable individual cortical parcellations across multiple sessions with different scanners. Moreover, it keeps highly group consistence as well as individual-specific variations. Given the reliable inter-subject variabilities from BAI-Net parcellation, their functional connectome showed higher associations with individual behavioral scores in a cognitive battery, compared to the individualized methods. The presented model has potential applications in locating more individual-specific regions in the diagnosis and treatment of neurological disorders, rather than group-registered method in terms of cortical morphology. And its methodology could be applied to individualize many population brain atlases.

show abstract

“…The connectivity profile can be defined in terms of the white matter pathway represented by tractography through diffusion magnetic resonance imaging (MRI), or in terms of the temporal coupling between spontaneous fluctuations of resting-state functional MRI (rfMRI) signal. Under the proposal of Passingham et al (2002), previous studies have utilized structural connectivity (Johansen-Berg et al 2004;Tomassini et al 2007;Beckmann et al 2009;Saygin et al 2011a) or functional connectivity (Cohen et al 2008;Gordon et al 2016) to characterize the boundary of functionally distinct brain regions, or have utilized structural connectivity (Saygin et al 2011b;Osher et al 2016;Saygin et al 2016;Wu et al 2020) or functional connectivity (Tavor et al 2016;Parker Jones et al 2017) to predict the functional activation information of brain regions at various task states.…”

Section: Introductionmentioning

confidence: 99%

Multi-hops functional connectivity improves individual prediction of fusiform face activation via a graph neural network

Wu¹,

Feng

2020

Preprint

Self Cite

View full text Add to dashboard Cite

Brain connectivity plays an important role in determining the brain region's function. Previous researchers proposed that the brain region's function is characterized by that region's input and output connectivity profiles. Following this proposal, numerous studies have investigated the relationship between connectivity and function. However, based on a preliminary analysis, this proposal is deficient in explaining individual differences in the brain region's function. To overcome this problem, we proposed that a brain region's function is characterized by that region's multi-hops connectivity profile. To test this proposal, we used multi-hops functional connectivity to predict the individual face response of the right fusiform face area (rFFA) via a multi-layers graph neural network and showed that the prediction performance is essentially improved. Results also indicated that the 2-layers graph neural network is the best in characterizing rFFA's face response and revealed a hierarchical network for the face processing of rFFA.

show abstract

Hierarchy of connectivity-function relationship of the human cortex revealed through predicting activity across functional domains

Cited by 8 publications

References 60 publications

Unravelling the Neural Basis of Spatial Delusions After Stroke

Unravelling the Neural Basis of Spatial Delusions After Stroke

BAI-Net: Individualized Anatomical Cerebral Cartography using Graph Neural Network

Multi-hops functional connectivity improves individual prediction of fusiform face activation via a graph neural network

Contact Info

Product

Resources

About