Filters








279 Hits in 3.6 sec

Intrinsic Patch-Based Cortical Anatomical Parcellation Using Graph Convolutional Neural Network on Surface Manifold [chapter]

Zhengwang Wu, Fenqiang Zhao, Jing Xia, Li Wang, Weili Lin, John H. Gilmore, Gang Li, Dinggang Shen
2019 Lecture Notes in Computer Science  
ability of the graph convolutional neural networks.  ...  Finally, parcellation labels on the spherical surface are mapped back to the original cortical surface.  ...  Figure 2(a) shows the typical sampling positions on a cortical surface. Convolution on Graph Convolutional neural networks (CNNs) are powerful methods for learning highly nonlinear mappings.  ... 
doi:10.1007/978-3-030-32248-9_55 pmid:32128522 pmcid:PMC7052684 fatcat:7f7zpnrxlbcmdiklzutrcpj5aa

Spherical U-Net on Cortical Surfaces: Methods and Applications [article]

Fenqiang Zhao, Shunren Xia, Zhengwang Wu, Dingna Duan, Li Wang, Weili Lin, John H Gilmore, Dinggang Shen, Gang Li
2019 arXiv   pre-print
Convolutional Neural Networks (CNNs) have been providing the state-of-the-art performance for learning-related problems involving 2D/3D images in Euclidean space.  ...  We then apply the Spherical U-Net to two challenging and neuroscientifically important tasks in infant brains: cortical surface parcellation and cortical attribute map development prediction.  ...  Introduction Convolutional Neural Networks (CNNs) based deep learning methods have been providing the state-of-the-art performance for a variety of tasks in computer vision and biomedical image analysis  ... 
arXiv:1904.00906v1 fatcat:dwa2j6kwang6te4igp6tjk2mcq

Spherical U-Net on Cortical Surfaces: Methods and Applications [chapter]

Fenqiang Zhao, Shunren Xia, Zhengwang Wu, Dingna Duan, Li Wang, Weili Lin, John H. Gilmore, Dinggang Shen, Gang Li
2019 Lecture Notes in Computer Science  
Convolutional Neural Networks (CNNs) have been providing the state-of-the-art performance for learning-related problems involving 2D/3D images in Euclidean space.  ...  We then apply the Spherical U-Net to two challenging and neuroscientifically important tasks in infant brains: cortical surface parcellation and cortical attribute map development prediction.  ...  To validate our proposed network, we demonstrate the capability and efficiency of the Spherical U-Net architecture on two challenging tasks in infant brains: cortical surface parcellation, which is a vertex-wise  ... 
doi:10.1007/978-3-030-20351-1_67 pmid:32180666 pmcid:PMC7074928 fatcat:iwt4kgh5sfhbdiqiycto5ypzxe

Cortical surface registration using unsupervised learning [article]

Jieyu Cheng, Adrian V. Dalca, Bruce Fischl, Lilla Zollei
2020 arXiv   pre-print
Recently, convolutional neural networks (CNNs) have demonstrated the potential to dramatically speed up volumetric registration.  ...  A conventional solution is to use a spherical representation of surface properties and perform registration by aligning cortical folding patterns in that space.  ...  Most existing work, however, focuses on the construction of spherical convolutional kernels and, to the best of our knowledge, neural networks have not yet been extended to surface registration.  ... 
arXiv:2004.04617v2 fatcat:kexmfiiu5bhvnig546qbbsuip4

Cortical surface registration using unsupervised learning

Jieyu Cheng, Adrian V. Dalca, Bruce Fischl, Lilla Zöllei
2020 NeuroImage  
Recently, convolutional neural networks (CNNs) have demonstrated the potential to dramatically speed up volumetric registration.  ...  A conventional solution is to use a spherical representation of surface properties and perform registration by aligning cortical folding patterns in that space.  ...  Most existing work, however, focuses on the construction of spherical convolutional kernels and, to the best of our knowledge, neural networks have not yet been extended to surface registration.  ... 
doi:10.1016/j.neuroimage.2020.117161 pmid:32702486 pmcid:PMC7784120 fatcat:7eqo3zodxbgg5j7lmf4z2qkyme

Graph Convolutions on Spectral Embeddings: Learning of Cortical Surface Data [article]

Karthik Gopinath, Christian Desrosiers, Herve Lombaert
2018 arXiv   pre-print
For instance, the widely used FreeSurfer takes about 3 hours to parcellate brain surfaces on a standard machine.  ...  The analysis of surface data is, however, challenging due to the high variability of the cortical geometry.  ...  Finally, a geometric convolutional neural network (CNN) is used to map input features, corresponding to the spectral coordinates and sulcal depth of brain graph nodes, to a labeled graph.  ... 
arXiv:1803.10336v1 fatcat:mvcxhzeuqzcvdolw3j3n4t3lxm

Non-Euclidean, convolutional learning on cortical brain surfaces

Mahmoud Mostapha, SunHyung Kim, Guorong Wu, Leo Zsembik, Stephen Pizer, Martin Styner
2018 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018)  
To solve this issue, this paper proposes a novel data-driven approach by extending convolutional neural networks (CNN) for use on non-Euclidean manifolds such as cortical surfaces.  ...  In recent years there have been many studies indicating that multiple cortical features, extracted at each surface vertex, are promising in the detection of various neurodevelopmental and neurodegenerative  ...  Convolution on Cortical Surfaces Convolution of data living on cortical surfaces is defined as a correlation with a surface kernel that is used to extract corresponding patches on the manifold.  ... 
doi:10.1109/isbi.2018.8363631 pmid:30364770 pmcid:PMC6197818 dblp:conf/isbi/MostaphaK0ZPS18 fatcat:337bzv5fyzczrcvdkylduedx7e

Geometric Deep Learning of the Human Connectome Project Multimodal Cortical Parcellation [article]

Logan Z J Williams, Abdulah Fawaz, Matthew F Glasser, David Edwards, Emma C Robinson
2021 bioRxiv   pre-print
While many cortical parcellation schemes have been proposed, few attempt to model inter-subject variability.  ...  In this paper, we benchmark and ensemble four different geometric deep learning models on the task of learning the Human Connectome Project (HCP) multimodal cortical parcellation.  ...  To this end, we consider convolutional neural networks (CNNs), which have proven state-of-the-art for many 2D and 3D medical imaging tasks [16, 3] .  ... 
doi:10.1101/2021.08.18.456790 fatcat:oia2azyo2zbjha2kotfbph6rau

Learning Cortical Parcellations Using Graph Neural Networks

Kristian M. Eschenburg, Thomas J. Grabowski, David R. Haynor
2021 Frontiers in Neuroscience  
Convolutional neural networks have been particularly useful for analyzing MRI data due to the regularly sampled spatial and temporal nature of the 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.  ...  Glasser for making subject-level Human Connectome Project multi-modal parcellations available for this analysis, and for his helpful and extensive comments on a draft version of the manuscript.  ... 
doi:10.3389/fnins.2021.797500 pmid:35002611 pmcid:PMC8739886 fatcat:u3agq3yvnrg6bpvflbwoa3ptre

Learnable Pooling in Graph Convolution Networks for Brain Surface Analysis [article]

Karthik Gopinath, Christian Desrosiers, Herve Lombaert
2019 arXiv   pre-print
Recent advances in machine learning have enabled the use of neural networks for non-Euclidean spaces.  ...  Our experiments demonstrate the superiority of our learnable pooling approach compared to other pooling techniques for graph convolution networks, with results improving the state-of-the-art in brain surface  ...  Fig. 2 . 2 An overview of the proposed graph convolution network for subject-specific cortical surface analysis.  ... 
arXiv:1911.10129v1 fatcat:gq6eekjux5b37jwekcrjjdieom

Benchmarking Geometric Deep Learning for Cortical Segmentation and Neurodevelopmental Phenotype Prediction [article]

Abdulah Fawaz, Logan Zane John Williams, Amir Alansary, Cher Bass, Karthik Gopinath, Mariana Da Silva, Simon Dahan, Christopher L Adamson, Bonnie Alexander, Deanne Thompson, Gareth Ball, Christian Desrosiers (+4 others)
2021 bioRxiv   pre-print
The emerging field of geometric deep learning extends the application of convolutional neural networks to irregular domains such as graphs, meshes and surfaces.  ...  Several recent studies have explored the potential for using these techniques to analyse and segment the cortical surface.  ...  Geometric deep learning (gDL) is the branch of research tasked at adapting convolutional neural networks (CNNs) to irregular domains such as sur-2 faces, meshes and graphs .  ... 
doi:10.1101/2021.12.01.470730 fatcat:szpev7wqenax7mds53ncg5stte

Graph-Based Deep Learning for Medical Diagnosis and Analysis: Past, Present and Future

David Ahmedt-Aristizabal, Mohammad Ali Armin, Simon Denman, Clinton Fookes, Lars Petersson
2021 Sensors  
As such, graph neural networks have attracted significant attention by exploiting implicit information that resides in a biological system, with interacting nodes connected by edges whose weights can be  ...  The spherical UNet is efficient in learning useful features to predict cortical surface parcellation and cortical attribute map development.  ...  Spherical UNet architecture. The output surface is a cortical parcellation map or a cortical attribute map, and the blue boxes reflect feature maps in spherical space.  ... 
doi:10.3390/s21144758 fatcat:jytyt4u2pjgvhnhcto3vcvd3a4

Representation Learning of Resting State fMRI with Variational Autoencoder

Jung-Hoon Kim, Yizhen Zhang, Kuan Han, Zheyu Wen, Minkyu Choi, Zhongming Liu
2021 NeuroImage  
The latent variables reflect the principal gradients of the latent trajectory and drive activity changes in cortical networks.  ...  After being trained with large data from the Human Connectome Project, the model has learned to represent and generate patterns of cortical activity and connectivity using latent variables.  ...  Geometric reformatting We converted the rsfMRI data from 3-D cortical surfaces to 2-D grids in order to structure the rsfMRI pattern as an image to ease the application of convolutional neural networks  ... 
doi:10.1016/j.neuroimage.2021.118423 pmid:34303794 fatcat:ctec5vqvazgsdda5jz42nhitva

Relations Between the Geometry of Cortical Gyrification and White-Matter Network Architecture

James A. Henderson, Peter A. Robinson
2014 Brain Connectivity  
However, the impact on network structure of convoluted cortical geometry is unknown.  ...  A geometrically based network model of cortico-cortical white matter connectivity is used in combination with DSI data to show that white matter cortical network architecture is founded on a homogeneous  ...  Figure 1 : 1 Parcellation of the cortical surface. (a) The cortical surface mesh parcellated into 989 nodes.  ... 
doi:10.1089/brain.2013.0183 pmid:24437717 fatcat:qptgm4t2lbdgpn2wdkbd7trn7y

Cartography and Connectomes

David C. Van Essen
2013 Neuron  
Here, the focus is on analyses that use surface reconstructions of individual subjects followed by registration to a surface-based atlas in order to cope with the complexity of human cortical convolutions  ...  A key to circumventing this difficulty is to use surface-based representations that respect the sheet-like topology of cortical structures.  ... 
doi:10.1016/j.neuron.2013.10.027 pmid:24183027 pmcid:PMC3855872 fatcat:nsrernbfkvctlbgaigmef7bedy
« Previous Showing results 1 — 15 out of 279 results