scholarly journals Learning Cortical Parcellations Using Graph Neural Networks

2021 ◽  
Vol 15 ◽  
Author(s):  
Kristian M. Eschenburg ◽  
Thomas J. Grabowski ◽  
David R. Haynor
Keyword(s):  
Neural Networks ◽  
Resting State ◽  
Network Architecture ◽  
Brain Connectivity ◽  
Model Performance ◽  
Disease Diagnosis ◽  
General Purpose ◽  
Tissue Segmentation ◽  
Graph Neural Networks ◽  
Resting State Connectivity

Deep learning has been applied to magnetic resonance imaging (MRI) for a variety of purposes, ranging from the acceleration of image acquisition and image denoising to tissue segmentation and disease diagnosis. Convolutional neural networks have been particularly useful for analyzing MRI data due to the regularly sampled spatial and temporal nature of the data. However, advances in the field of brain imaging have led to network- and surface-based analyses that are often better represented in the graph domain. In this analysis, we propose a general purpose cortical segmentation method that, given resting-state connectivity features readily computed during conventional MRI pre-processing and a set of corresponding training labels, can generate cortical parcellations for new MRI data. We applied recent advances in the field of graph neural networks to the problem of cortical surface segmentation, using resting-state connectivity to learn discrete maps of the human neocortex. We found that graph neural networks accurately learn low-dimensional representations of functional brain connectivity that can be naturally extended to map the cortices of new datasets. After optimizing over algorithm type, network architecture, and training features, our approach yielded mean classification accuracies of 79.91% relative to a previously published parcellation. We describe how some hyperparameter choices including training and testing data duration, network architecture, and algorithm choice affect model performance.

scholarly journals SketchGNN: Semantic Sketch Segmentation with Graph Neural Networks

2021 ◽  
Vol 40 (3) ◽  
pp. 1-13
Author(s):  
Lumin Yang ◽  
Jiajie Zhuang ◽  
Hongbo Fu ◽  
Xiangzhi Wei ◽  
Kun Zhou ◽  
...  
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Network Architecture ◽  
Large Scale ◽  
State Of The Art ◽  
Semantic Segmentation ◽  
Structure Information ◽  
Graph Neural Networks ◽  
Node Labels ◽  
Point Level

We introduce SketchGNN , a convolutional graph neural network for semantic segmentation and labeling of freehand vector sketches. We treat an input stroke-based sketch as a graph with nodes representing the sampled points along input strokes and edges encoding the stroke structure information. To predict the per-node labels, our SketchGNN uses graph convolution and a static-dynamic branching network architecture to extract the features at three levels, i.e., point-level, stroke-level, and sketch-level. SketchGNN significantly improves the accuracy of the state-of-the-art methods for semantic sketch segmentation (by 11.2% in the pixel-based metric and 18.2% in the component-based metric over a large-scale challenging SPG dataset) and has magnitudes fewer parameters than both image-based and sequence-based methods.

scholarly journals On the Effectiveness of Bayesian AutoML methods for Physics Emulators

2020 ◽  
Author(s):  
Peetak Mitra ◽  
Niccolò Dal Santo ◽  
Majid Haghshenas ◽  
Shounak Mitra ◽  
Conor Daly ◽  
...  
Keyword(s):  
Neural Networks ◽  
Network Architecture ◽  
Model Performance ◽  
Parameter Tuning ◽  
Bayesian Optimization ◽  
Eddy Simulation ◽  
Large Eddy ◽  
Frame Work ◽  
Network Weight ◽  
And Function

The adoption of Machine Learning (ML) for building emulators for complex physical processes has seen an exponential rise in the recent years. While neural networks are good function approximators, optimizing the hyper-parameters of the network to reach a global minimum is not trivial, and often needs human knowl- edge and expertise. In this light, automatic ML or autoML methods have gained large interest as they automate the process of network hyper-parameter tuning. In addition, Neural Architecture Search (NAS) has shown promising outcomes for improving model performance. While autoML methods have grown in popularity for image, text and other applications, their effectiveness for high-dimensional, complex scientific datasets remains to be investigated. In this work, a data driven emulator for turbulence closure terms in the context of Large Eddy Simulation (LES) models is trained using Artificial Neural Networks and an autoML frame- work based on Bayesian Optimization, incorporating priors to jointly optimize the hyper-parameters as well as conduct a full neural network architecture search to converge to a global minima, is proposed. Additionally, we compare the effect of using different network weight initializations and optimizers such as ADAM, SGDM and RMSProp, to explore the best performing setting. Weight and function space similarities during the optimization trajectory are investigated, and critical differences in the learning process evolution are noted and compared to theory. We observe ADAM optimizer and Glorot initialization consistently performs better, while RMSProp outperforms SGDM as the latter appears to have been stuck at a local minima. Therefore, this autoML BayesOpt framework provides a means to choose the best hyper-parameter settings for a given dataset.

scholarly journals Towards Heterogeneous Multi-Agent Reinforcement Learning with Graph Neural Networks

2020 ◽  
Author(s):  
Douglas Meneghetti ◽  
Reinaldo Bianchi
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Network Architecture ◽  
Communication Channels ◽  
Neural Network Architecture ◽  
Graph Representations ◽  
Labeled Graph ◽  
Multiple Agent ◽  
Multi Agent ◽  
Graph Neural Networks

This work proposes a neural network architecture that learns policies for multiple agent classes in a heterogeneous multi-agent reinforcement setting. The proposed network uses directed labeled graph representations for states, encodes feature vectors of different sizes for different entity classes, uses relational graph convolution layers to model different communication channels between entity types and learns distinct policies for different agent classes, sharing parameters wherever possible. Results have shown that specializing the communication channels between entity classes is a promising step to achieve higher performance in environments composed of heterogeneous entities.

scholarly journals Biological network analysis with deep learning

2020 ◽  
Author(s):  
Giulia Muzio ◽  
Leslie O’Bray ◽  
Karsten Borgwardt
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Biological Networks ◽  
Protein Function ◽  
Regulatory Networks ◽  
Biological Network ◽  
Gene Interaction ◽  
Disease Diagnosis ◽  
Interaction Prediction ◽  
Graph Neural Networks

Abstract Recent advancements in experimental high-throughput technologies have expanded the availability and quantity of molecular data in biology. Given the importance of interactions in biological processes, such as the interactions between proteins or the bonds within a chemical compound, this data is often represented in the form of a biological network. The rise of this data has created a need for new computational tools to analyze networks. One major trend in the field is to use deep learning for this goal and, more specifically, to use methods that work with networks, the so-called graph neural networks (GNNs). In this article, we describe biological networks and review the principles and underlying algorithms of GNNs. We then discuss domains in bioinformatics in which graph neural networks are frequently being applied at the moment, such as protein function prediction, protein–protein interaction prediction and in silico drug discovery and development. Finally, we highlight application areas such as gene regulatory networks and disease diagnosis where deep learning is emerging as a new tool to answer classic questions like gene interaction prediction and automatic disease prediction from data.

scholarly journals Individualized Spatial Network Predictions Using Siamese Convolutional Neural Networks: A Resting-State fMRI Study of over 11,000 Unaffected Individuals

2021 ◽  
Author(s):  
Reihaneh Hassanzadeh ◽  
Rogers F. Silva ◽  
Anees Abrol ◽  
Mustafa Salman ◽  
Anna Bonkhoff ◽  
...  
Keyword(s):  
Neural Networks ◽  
Spatial Variability ◽  
Resting State ◽  
Life Experience ◽  
Predictive Accuracy ◽  
Brain Connectivity ◽  
Brain Anatomy ◽  
Spatial Networks ◽  
Domain Level ◽  
Spatial Maps

Individuals can be characterized in a population according to their brain measurements and activity, given the inter-subject variability in brain anatomy, structure-function relationships, or life experience. Many neuroimaging studies have demonstrated the potential of functional network connectivity patterns estimated from resting functional magnetic resonance imaging (fMRI) to discriminate groups and predict information about individual subjects. However, the predictive signal present in the spatial heterogeneity of brain connectivity networks is yet to be extensively studied. In this study, we investigate, for the first time, the use of pairwise-relationships between resting-state independent  spatial maps  to characterize individuals. To do this, we develop a deep Siamese framework comprising three-dimensional convolution neural networks for contrastive learning based on individual-level spatial maps estimated via a fully automated fMRI independent component analysis approach. The proposed framework evaluates whether pairs of spatial networks (e.g., visual network and auditory network) are capable of subject identification and assesses the spatial variability in different network pairs' predictive power in an extensive whole-brain analysis. Our analysis on nearly 12,000 unaffected individuals from the UK Biobank study demonstrates that the proposed approach can discriminate subjects with an accuracy of up to 88% for a single network pair on the test set (best model, after several runs), and 82% average accuracy at the subcortical domain level, notably the highest average domain level accuracy attained. Further investigation of our network's learned features revealed a higher spatial variability in predictive accuracy among younger brains and significantly higher  discriminative power among males. In sum, the relationship among spatial networks appears to be both informative and discriminative of individuals and should be studied further as putative brain-based biomarkers.

scholarly journals Neural Networks Implicated in Autobiographical Memory Training

2021 ◽  
Author(s):  
Dragoş Cȋrneci ◽  
Mihaela Onu ◽  
Claudiu C. Papasteri ◽  
Dana Georgescu ◽  
Catalina Poalelungi ◽  
...  
Keyword(s):  
Neural Networks ◽  
Autobiographical Memory ◽  
Resting State ◽  
Immune Modulation ◽  
Neural Substrates ◽  
Memory Training ◽  
Sensorimotor Network ◽  
Voluntary Recall ◽  
High Level ◽  
Resting State Connectivity

Abstract Training of autobiographical memory has been proposed as intervention to improve cognitive functions. The neural substrates for such improvements are poorly understood. Several brain networks have been previously linked to autobiographical recollections, including the default mode network (DMN) and the sensorimotor network. Here we tested the hypothesis that different neural networks support distinct aspects of memory improvement in response to training on a group of 59 subjects. We found that memory training using olfactory cues increases resting-state intra-network DMN connectivity, and this associates with improved recollection of cue-specific memories. On the contrary, training decreased resting-state connectivity within the sensorimotor network, a decrease that correlated with improved ability for voluntary recall. Moreover, only the decrease in sensorimotor connectivity associated with the training-induced decrease in the TNFα factor, an immune modulation previously linked to improved cognitive performance. We identified functional and biochemical factors that associate with distinct memory processes improved by autobiographical training. Pathways which connect autobiographical memory to both high level cognition and somatic physiology are discussed.

scholarly journals Resting-state connectivity studies as a marker of the acute and delayed effects of subanaesthetic ketamine administration in healthy and depressed individuals: A systematic review

2021 ◽  
Vol 5 ◽  
pp. 239821282110554
Author(s):  
Vasileia Kotoula ◽  
Toby Webster ◽  
James Stone ◽  
Mitul A Mehta
Keyword(s):  
Resting State ◽  
Emotional Regulation ◽  
Brain Connectivity ◽  
Brain Activity ◽  
Reward Processing ◽  
Brain Areas ◽  
Delayed Effects ◽  
Treatment Naïve ◽  
Resting State Connectivity ◽  
The Brain

Acute ketamine administration has been widely used in neuroimaging research to mimic psychosis-like symptoms. Within the last two decades, ketamine has also emerged as a potent, fast-acting antidepressant. The delayed effects of the drug, observed 2–48 h after a single infusion, are associated with marked improvements in depressive symptoms. At the systems’ level, several studies have investigated the acute ketamine effects on brain activity and connectivity; however, several questions remain unanswered around the brain changes that accompany the drug’s antidepressant effects and how these changes relate to the brain areas that appear with altered function and connectivity in depression. This review aims to address some of these questions by focusing on resting-state brain connectivity. We summarise the studies that have examined connectivity changes in treatment-naïve, depressed individuals and those studies that have looked at the acute and delayed effects of ketamine in healthy and depressed volunteers. We conclude that brain areas that are important for emotional regulation and reward processing appear with altered connectivity in depression whereas the default mode network presents with increased connectivity in depressed individuals compared to healthy controls. This finding, however, is not as prominent as the literature often assumes. Acute ketamine administration causes an increase in brain connectivity in healthy volunteers. The delayed effects of ketamine on brain connectivity vary in direction and appear to be consistent with the drug normalising the changes observed in depression. The limited number of studies however, as well as the different approaches for resting-state connectivity analysis make it very difficult to draw firm conclusions and highlight the importance of data sharing and larger future studies.

scholarly journals Predicting Cognitive Scores With Graph Neural Networks Through Sample Selection Learning

2021 ◽  
Author(s):  
Martin Hanik ◽  
Mehmet Arif Demirtaş ◽  
Mohammed Amine Gharsallaoui ◽  
Islem Rekik
Keyword(s):  
Neural Networks ◽  
Brain Connectivity ◽  
Sample Selection ◽  
Target Prediction ◽  
Autism Spectrum ◽  
Selection Method ◽  
Brain Connectome ◽  
Training Samples ◽  
Iq Scores ◽  
Graph Neural Networks

Abstract Analyzing the relation between intelligence and neural activity is of the utmost importance in understanding the working principles of the human brain in health and disease. In existing literature, functional brain connectomes have been used successfully to predict cognitive measures such as intelligence quotient (IQ) scores in both healthy and disordered cohorts using machine learning models. However, existing methods resort to flattening the brain connectome (i.e., graph) through vectorization which overlooks its topological properties. To address this limitation and inspired from the emerging graph neural networks (GNNs), we design a novel regression GNN model (namely RegGNN) for predicting IQ scores from brain connectivity. On top of that, we introduce a novel, fully modular sample selection method to select the best samples to learn from for our target prediction task. However, since such deep learning architectures are computationally expensive to train, we further propose a \emph{learning-based sample selection} method that learns how to choose the training samples with the highest expected predictive power on unseen samples. For this, we capitalize on the fact that connectomes (i.e., their adjacency matrices) lie in the symmetric positive definite (SPD) matrix cone. Our results on full-scale and verbal IQ prediction outperforms comparison methods in autism spectrum disorder cohorts and achieves a competitive performance for neurotypical subjects using 3-fold cross-validation. Furthermore, we show that our sample selection approach generalizes to other learning-based methods, which shows its usefulness beyond our GNN architecture.

scholarly journals The Protective Effects of Supportive Parenting on the Relationship Between Adolescent Poverty and Resting-State Functional Brain Connectivity During Adulthood

2019 ◽  
Vol 30 (7) ◽  
pp. 1040-1049 ◽  
Author(s):  
Gene H. Brody ◽  
Tianyi Yu ◽  
Robin Nusslock ◽  
Allen W. Barton ◽  
Gregory E. Miller ◽  
...  
Keyword(s):  
Neural Networks ◽  
Emotion Regulation ◽  
Resting State ◽  
Brain Connectivity ◽  
Protective Effects ◽  
Growing Up ◽  
Resting State Functional Connectivity ◽  
Cognition And Emotion ◽  
Supportive Parenting ◽  
Poverty Status

Children growing up in poverty are vulnerable to negative changes in the developing brain; however, these outcomes vary widely. We tested the hypothesis that receipt of supportive parenting would offset the association between living in poverty during adolescence and the connectivity of neural networks that support cognition and emotion regulation during young adulthood. In a sample of African American youths ( N = 119) living in the rural South, poverty status and receipt of supportive parenting were assessed when youths were 11 to 13 and 16 to 18 years old. At age 25, resting-state functional connectivity of the central-executive and emotion-regulation neural networks was assessed using functional MRI. The results revealed that more years spent living in poverty presaged less connectivity in both neural networks among young adults who received low levels of supportive parenting but not among those who received high levels of such parenting.

scholarly journals Rumor Detection Based on SAGNN: Simplified Aggregation Graph Neural Networks

2021 ◽  
Vol 3 (1) ◽  
pp. 84-94
Author(s):  
Liang Zhang ◽  
Jingqun Li ◽  
Bin Zhou ◽  
Yan Jia
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Network Architecture ◽  
Network Architectures ◽  
Neural Network Architecture ◽  
Convolutional Networks ◽  
Effective Manner ◽  
Deep Learning Network ◽  
Graph Neural Networks ◽  
Rumor Detection

Identifying fake news on media has been an important issue. This is especially true considering the wide spread of rumors on popular social networks such as Twitter. Various kinds of techniques have been proposed for automatic rumor detection. In this work, we study the application of graph neural networks for rumor classification at a lower level, instead of applying existing neural network architectures to detect rumors. The responses to true rumors and false rumors display distinct characteristics. This suggests that it is essential to capture such interactions in an effective manner for a deep learning network to achieve better rumor detection performance. To this end we present a simplified aggregation graph neural network architecture. Experiments on publicly available Twitter datasets demonstrate that the proposed network has performance on a par with or even better than that of state-of-the-art graph convolutional networks, while significantly reducing the computational complexity.

Sign in / Sign up

Export Citation Format

Share Document