Combining Multiscale Diffusion Kernels for Learning the Structural and Functional Brain Connectivity

Sg, Surampudi; Naik, Shruti; Sharma, Avinash; Bapi, Raju S.; Roy, Dipanjan

doi:10.1101/078766

Cited by 4 publications

(8 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, this model uses one global parameter across all the subjects, and the hypothesis of a single scale best-fitting kernel across subjects is not tenable. Surampudi et al [6] observed that the combination of multiple diffusion scales exhibits scale-dependent relationships among various regions of interest 115 (ROIs), and these multi-scale diffusion kernels can capture reaction-diffusion systems operating on a fixed underlying connectome (SC). However, multiple diffusion kernels were not sufficient to explain the self-organizing resting-state patterns found in FC.…”

Section: Graph-theoretic Modeling Using Linear Modelsmentioning

confidence: 99%

“…Similarly, in [6], the mixing coefficients are subsequently learned while solving an optimization formulation as:…”

Section: Relation To Reaction Diffusion Phenomenonmentioning

confidence: 99%

“…One of the seminal works in this direction by [3] formulated a linear model that considers diffusion of regional brain activity over the graph topology by choosing a single optimal diffusion kernel. Later, [6,7] utilized multiple diffusion kernels for learning the SC-FC mapping and demonstrated the superiority of using multiple diffusion kernels. The idea of multiple diffusion kernels formulation is specifically interesting and demonstrates the integration of multiple kernels within the same machine learning model.…”

Section: Introductionmentioning

confidence: 99%

“…The copyright holder for this preprint this version posted August 13, 2021. ; https://doi.org/10.1101/2021.08.12.456134 doi: bioRxiv preprint The study of the relationship between structural connectivity and functional connectivity and how the functional activity of the brain is generated from the anatomical structure has been a major research topic in the field of cognitive neuroscience. Several methods have been proposed to explore the mapping between SC-FC including, whole brain computational models [22,23], simple linear diffusion models [3] as well as complex non-linear models [25,27], and linear multi-scale diffusion models [6,7]. The whole brain computational models have been used as powerful tools to understand the relationship between structural and functional brain connectivity by linking brain function with its physiological underpinnings.…”

mentioning

confidence: 99%

“…Abdelnour et al [3] introduced a graph-based model with a linear single scale diffusion kernel at an optimal scale over the structural graph topology (SC) to map FC. However, Surampudi et al [6] showed that single kernel models do not generalise to a larger cohort and demonstrated that FC can be decomposed into multiple diffusion kernels with subject non-specific combination coefficients. Further, the MKL framework, proposed by Surampudi et al [7], revealed that the combination of multiple diffusion kernels was not sufficient to explain the self-organizing resting-state patterns found in FC and hence necessitated the use of additional explanatory 490 parameters.…”

mentioning

confidence: 99%

See 4 more Smart Citations

A-GHN: Attention-based Fusion of Multiple GraphHeat Networks for Structural to Functional Brain Mapping

Oota¹,

Yadav²,

Dash³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Over the last decade, there has been growing interest in learning the mapping from structural connectivity (SC) to functional connectivity (FC) of the brain. The spontaneous fluctuations of the brain activity during the resting-state as captured by functional MRI (rsfMRI) contain rich non-stationary dynamics over a relatively fixed structural connectome. Among the modeling approaches, graph diffusion-based methods with single and multiple diffusion kernels approximating static or dynamic functional connectivity have shown promise in predicting the FC given the SC. However, these methods are computationally expensive, not scalable, and fail to capture the complex dynamics underlying the whole process. Recently, deep learning methods such as GraphHeat networks along with graph diffusion have been shown to handle complex relational structures while preserving global information. In this paper, we propose a novel attention-based fusion of multiple GraphHeat networks (A-GHN) for mapping SC-FC. A-GHN enables us to model multiple heat kernel diffusion over the brain graph for approximating the complex Reaction Diffusion phenomenon. We argue that the proposed deep learning method overcomes the scalability and computational inefficiency issues but can still learn the SC-FC mapping successfully. Training and testing were done using the rsfMRI data of 100 participants from the human connectome project (HCP), and the results establish the viability of the proposed model. Furthermore, experiments demonstrate that A-GHN outperforms the existing methods in learning the complex nature of human brain function.

show abstract

Section: Graph-theoretic Modeling Using Linear Modelsmentioning

confidence: 99%

“…Similarly, in [6], the mixing coefficients are subsequently learned while solving an optimization formulation as:…”

Section: Relation To Reaction Diffusion Phenomenonmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

See 3 more Smart Citations

A-GHN: Attention-based Fusion of Multiple GraphHeat Networks for Structural to Functional Brain Mapping

Oota¹,

Yadav²,

Dash³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

Resting state dynamics meets anatomical structure: Temporal multiple kernel learning (tMKL) model

et al. 2019

View full text Add to dashboard Cite

Attention-based fusion of multiple graphheat networks for structural to functional brain mapping

Oota,

Yadav,

Dash

et al. 2024

Sci Rep

View full text Add to dashboard Cite

Over the last decade, there has been growing interest in learning the mapping from structural connectivity (SC) to functional connectivity (FC) of the brain. The spontaneous fluctuations of the brain activity during the resting-state as captured by functional MRI (rsfMRI) contain rich non-stationary dynamics over a relatively fixed structural connectome. Among the modeling approaches, graph diffusion-based methods with single and multiple diffusion kernels approximating static or dynamic functional connectivity have shown promise in predicting the FC given the SC. However, these methods are computationally expensive, not scalable, and fail to capture the complex dynamics underlying the whole process. Recently, deep learning methods such as GraphHeat networks and graph diffusion have been shown to handle complex relational structures while preserving global information. In this paper, we propose a novel attention-based fusion of multiple GraphHeat networks (A-GHN) for mapping SC-FC. A-GHN enables us to model multiple heat kernel diffusion over the brain graph for approximating the complex Reaction Diffusion phenomenon. We argue that the proposed deep learning method overcomes the scalability and computational inefficiency issues but can still learn the SC-FC mapping successfully. Training and testing were done using the rsfMRI data of 1058 participants from the human connectome project (HCP), and the results establish the viability of the proposed model. On HCP data, we achieve a high Pearson correlation of 0.788 (Desikan-Killiany atlas with 87 regions) and 0.773 (AAL atlas with 86 regions). Furthermore, experiments demonstrate that A-GHN outperforms the existing methods in learning the complex nature of the structure-function relation of the human brain.

show abstract

Combining Multiscale Diffusion Kernels for Learning the Structural and Functional Brain Connectivity

Cited by 4 publications

References 19 publications

A-GHN: Attention-based Fusion of Multiple GraphHeat Networks for Structural to Functional Brain Mapping

A-GHN: Attention-based Fusion of Multiple GraphHeat Networks for Structural to Functional Brain Mapping

Resting state dynamics meets anatomical structure: Temporal multiple kernel learning (tMKL) model

Attention-based fusion of multiple graphheat networks for structural to functional brain mapping

Contact Info

Product

Resources

About