Recent searches

No recent searches

Search options

Only available when logged in.
med-mastodon.com is one of the many independent Mastodon servers you can use to participate in the fediverse.
Medical community on Mastodon

Administered by:

Server stats:

363
active users

med-mastodon.com: AboutProfiles directoryPrivacy policy

Mastodon: AboutGet the appKeyboard shortcutsView source codev4.4.2

#functionalconnectivity

0 posts0 participants0 posts today
Fabrice Tshimanga

Yesterday I attended a very interesting seminar with professors Alessandra Bertoldo, Alessandro Chiusi and Marco Zorzi from #unipd , from departments of information engineering, psychology, and neuroscience.
It was mostly about popularizing #fMRI and the clinical potential of such studies.
My mind was captured by two things, #effectiveconnectivity and the use of neural networks for #FunctionalConnectivity to symptoms mapping.
The core for me was: they are not talking about using deep learning, or the most apt deep learning architecture for the problems.
For EffectiveC., they were speaking about dynamical systems modeling (which is great!); for functional connectivity they cited convolutional autoencoders on the image or matrix of functional connectivity, which I really don't like unless number of channels and more importantly kernel dimension are discussed.
Overall, we are dealing with directed and undirected weighted graphs respectively, and we have architectures for those

Dominic Boutet

#arxivfeed :

"CI-GNN: A Granger Causality-Inspired Graph Neural Network for Interpretable Brain Network-Based Psychiatric Diagnosis"
arxiv.org/abs/2301.01642

arXiv.orgCI-GNN: A Granger Causality-Inspired Graph Neural Network for Interpretable Brain Network-Based Psychiatric DiagnosisThere is a recent trend to leverage the power of graph neural networks (GNNs) for brain-network based psychiatric diagnosis, which,in turn, also motivates an urgent need for psychiatrists to fully understand the decision behavior of the used GNNs. However, most of the existing GNN explainers are either post-hoc in which another interpretive model needs to be created to explain a well-trained GNN, or do not consider the causal relationship between the extracted explanation and the decision, such that the explanation itself contains spurious correlations and suffers from weak faithfulness. In this work, we propose a granger causality-inspired graph neural network (CI-GNN), a built-in interpretable model that is able to identify the most influential subgraph (i.e., functional connectivity within brain regions) that is causally related to the decision (e.g., major depressive disorder patients or healthy controls), without the training of an auxillary interpretive network. CI-GNN learns disentangled subgraph-level representations α and \b{eta} that encode, respectively, the causal and noncausal aspects of original graph under a graph variational autoencoder framework, regularized by a conditional mutual information (CMI) constraint. We theoretically justify the validity of the CMI regulation in capturing the causal relationship. We also empirically evaluate the performance of CI-GNN against three baseline GNNs and four state-of-the-art GNN explainers on synthetic data and two large-scale brain disease datasets. We observe that CI-GNN achieves the best performance in a wide range of metrics and provides more reliable and concise explanations which have clinical evidence.
#Neuroscience#Neuro#ComputationalNeuroscience
Dominic Boutet

#arxivfeed

"Mesoscopic patterns of functional connectivity alterations in autism by contrast subgraphs"
biorxiv.org/content/10.1101/20

bioRxivMesoscopic patterns of functional connectivity alterations in autism by contrast subgraphsDespite the breakthrough achievements in understanding structural and functional connectivity alterations that underlie autism spectrum disorder (ASD), the exact nature and type of such alterations are not yet clear due to conflicting reports of hyper-connectivity, hypo-connectivity, and --in some cases-- combinations of both. In this work, we approach the debate about hyper- vs hypo-connectivity in ASD using a novel network comparison technique designed to capture mesoscopic-scale differential structures. In particular, we build on recent algorithmic advances in the sparsification of functional connectivity matrices, in the extraction of contrast subgraphs, and in the computation of statistically significant maximal frequent itemsets, and develop a method to identify mesoscale structural subgraphs that are maximally dense and different in terms of connectivity levels between the different sets of networks. We apply our method to analyse brain networks of typically developed individuals and ASD patients across different developmental phases and find a set of altered cortical-subcortical circuits between healthy subjects and patients affected by ASD. Specifically, our analysis highlights in ASD patients a significantly larger number of functional connections among regions of the occipital cortex and between the left precuneus and the superior parietal gyrus. At the same time, reduced connectivity characterised the superior frontal gyrus and the temporal lobe regions. More importantly, we can simultaneously detect regions of the brain that show hyper and hypo-connectivity in ASD in children and adolescents, recapitulating within a single framework multiple previous separate observations. ### Competing Interest Statement The authors have declared no competing interest.
#Neuroscience#Neuro#Brain
TrendingLive feeds
About