Graph-Based Deep Learning for Prediction of Longitudinal Infant Diffusion MRI Data

Jaeil Kim, Yoonmi Hong, Geng Chen, Weili Lin, Pew Thian Yap, Dinggang Shen

Research output: Contribution to journalConference article

1 Citation (Scopus)

Abstract

Diffusion MRI affords great value for studying brain development, owing to its capability in assessing brain microstructure in association with myelination. With longitudinally acquired pediatric diffusion MRI data, one can chart the temporal evolution of microstructure and white matter connectivity. However, due to subject dropouts and unsuccessful scans, longitudinal datasets are often incomplete. In this work, we introduce a graph-based deep learning approach to predict diffusion MRI data. The relationships between sampling points in spatial domain (x-space) and diffusion wave-vector domain (q-space) are harnessed jointly (x-q space) in the form of a graph. We then implement a residual learning architecture with graph convolution filtering to learn longitudinal changes of diffusion MRI data along time. We evaluate the effectiveness of the spatial and angular components in data prediction. We also investigate the longitudinal trajectories in terms of diffusion scalars computed based on the predicted datasets.

Original languageEnglish
Pages (from-to)133-141
Number of pages9
JournalMathematics and Visualization
Issue number226249
DOIs
Publication statusPublished - 2019 Jan 1
Externally publishedYes
EventInternational Workshop on Computational Diffusion MRI, CDMRI 2018 held with International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2018 - Granada, Spain
Duration: 2018 Sep 202018 Sep 20

Fingerprint

Magnetic resonance imaging
Prediction
Graph in graph theory
Microstructure
Brain
Pediatrics
Drop out
Convolution
Chart
Learning
Deep learning
Connectivity
Filtering
Trajectories
Scalar
Trajectory
Sampling
Predict
Evaluate

Keywords

  • Brain development
  • Diffusion MRI
  • Graph convolution
  • Graph representation
  • Longitudinal prediction
  • Residual graph neural network

ASJC Scopus subject areas

  • Modelling and Simulation
  • Geometry and Topology
  • Computer Graphics and Computer-Aided Design
  • Applied Mathematics

Cite this

Graph-Based Deep Learning for Prediction of Longitudinal Infant Diffusion MRI Data. / Kim, Jaeil; Hong, Yoonmi; Chen, Geng; Lin, Weili; Yap, Pew Thian; Shen, Dinggang.

In: Mathematics and Visualization, No. 226249, 01.01.2019, p. 133-141.

Research output: Contribution to journalConference article

Kim, Jaeil ; Hong, Yoonmi ; Chen, Geng ; Lin, Weili ; Yap, Pew Thian ; Shen, Dinggang. / Graph-Based Deep Learning for Prediction of Longitudinal Infant Diffusion MRI Data. In: Mathematics and Visualization. 2019 ; No. 226249. pp. 133-141.
@article{8ba8ad7b570642b2ab8bfca0d5790f9d,
title = "Graph-Based Deep Learning for{\^A} Prediction of Longitudinal Infant Diffusion MRI Data",
abstract = "Diffusion MRI affords great value for studying brain development, owing to its capability in assessing brain microstructure in association with myelination. With longitudinally acquired pediatric diffusion MRI data, one can chart the temporal evolution of microstructure and white matter connectivity. However, due to subject dropouts and unsuccessful scans, longitudinal datasets are often incomplete. In this work, we introduce a graph-based deep learning approach to predict diffusion MRI data. The relationships between sampling points in spatial domain (x-space) and diffusion wave-vector domain (q-space) are harnessed jointly (x-q space) in the form of a graph. We then implement a residual learning architecture with graph convolution filtering to learn longitudinal changes of diffusion MRI data along time. We evaluate the effectiveness of the spatial and angular components in data prediction. We also investigate the longitudinal trajectories in terms of diffusion scalars computed based on the predicted datasets.",
keywords = "Brain development, Diffusion MRI, Graph convolution, Graph representation, Longitudinal prediction, Residual graph neural network",
author = "Jaeil Kim and Yoonmi Hong and Geng Chen and Weili Lin and Yap, {Pew Thian} and Dinggang Shen",
year = "2019",
month = "1",
day = "1",
doi = "10.1007/978-3-030-05831-9_11",
language = "English",
pages = "133--141",
journal = "Mathematics and Visualization",
issn = "1612-3786",
publisher = "Springer Science and Business Media Deutschland GmbH",
number = "226249",

}

TY - JOUR

T1 - Graph-Based Deep Learning for Prediction of Longitudinal Infant Diffusion MRI Data

AU - Kim, Jaeil

AU - Hong, Yoonmi

AU - Chen, Geng

AU - Lin, Weili

AU - Yap, Pew Thian

AU - Shen, Dinggang

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Diffusion MRI affords great value for studying brain development, owing to its capability in assessing brain microstructure in association with myelination. With longitudinally acquired pediatric diffusion MRI data, one can chart the temporal evolution of microstructure and white matter connectivity. However, due to subject dropouts and unsuccessful scans, longitudinal datasets are often incomplete. In this work, we introduce a graph-based deep learning approach to predict diffusion MRI data. The relationships between sampling points in spatial domain (x-space) and diffusion wave-vector domain (q-space) are harnessed jointly (x-q space) in the form of a graph. We then implement a residual learning architecture with graph convolution filtering to learn longitudinal changes of diffusion MRI data along time. We evaluate the effectiveness of the spatial and angular components in data prediction. We also investigate the longitudinal trajectories in terms of diffusion scalars computed based on the predicted datasets.

AB - Diffusion MRI affords great value for studying brain development, owing to its capability in assessing brain microstructure in association with myelination. With longitudinally acquired pediatric diffusion MRI data, one can chart the temporal evolution of microstructure and white matter connectivity. However, due to subject dropouts and unsuccessful scans, longitudinal datasets are often incomplete. In this work, we introduce a graph-based deep learning approach to predict diffusion MRI data. The relationships between sampling points in spatial domain (x-space) and diffusion wave-vector domain (q-space) are harnessed jointly (x-q space) in the form of a graph. We then implement a residual learning architecture with graph convolution filtering to learn longitudinal changes of diffusion MRI data along time. We evaluate the effectiveness of the spatial and angular components in data prediction. We also investigate the longitudinal trajectories in terms of diffusion scalars computed based on the predicted datasets.

KW - Brain development

KW - Diffusion MRI

KW - Graph convolution

KW - Graph representation

KW - Longitudinal prediction

KW - Residual graph neural network

UR - http://www.scopus.com/inward/record.url?scp=85066881735&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85066881735&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-05831-9_11

DO - 10.1007/978-3-030-05831-9_11

M3 - Conference article

AN - SCOPUS:85066881735

SP - 133

EP - 141

JO - Mathematics and Visualization

JF - Mathematics and Visualization

SN - 1612-3786

IS - 226249

ER -