Prediction of infant MRI appearance and anatomical structure evolution using sparse patch-based metamorphosis learning framework

Islem Rekik, Gang Li, Guorong Wu, Weili Lin, Dinggang Shen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Citations (Scopus)

Abstract

Magnetic resonance imaging (MRI) of pediatric brain provides invaluable information for early normal and abnormal brain development. Longitudinal neuroimaging has spanned various research works on examining infant brain development patterns. However, studies on predicting postnatal brain image evolution remain scarce, which is very challenging due to the dynamic tissue contrast change and even inversion in postnatal brains. In this paper, we unprecedentedly propose a dual image intensity and anatomical structure (label) prediction framework that nicely links the geodesic image metamorphosis model with sparse patch-based image representation, thereby defining spatiotemporal metamorphic patches encoding both image photometric and geometric deformation. In the training stage, we learn the 4D metamorphosis trajectories for each training subject. In the prediction stage, we define various strategies to sparsely represent each patch in the testing image using the training metamorphosis patches; while progressively incrementing the richness of the patch (from appearance-based to multimodal kinetic patches). We used the proposed framework to predict 6, 9 and 12-month brain MR image intensity and structure (white and gray matter maps) from 3 months in 10 infants. The proposed framework showed promising preliminary prediction results for the spatiotemporally complex, drastically changing brain images.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PublisherSpringer Verlag
Pages197-204
Number of pages8
Volume9467
ISBN (Print)9783319281933
DOIs
Publication statusPublished - 2015
Externally publishedYes
Event1st International Workshop on Patch-Based Techniques in Medical Imaging, Patch-MI 2015 - Munich, Germany
Duration: 2015 Oct 92015 Oct 9

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9467
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other1st International Workshop on Patch-Based Techniques in Medical Imaging, Patch-MI 2015
CountryGermany
CityMunich
Period15/10/915/10/9

Fingerprint

Magnetic Resonance Imaging
Magnetic resonance imaging
Patch
Brain
Prediction
Neuroimaging
Image Model
Pediatrics
Image Representation
Framework
Learning
Geodesic
Labels
Inversion
Encoding
Kinetics
Trajectories
Trajectory
Tissue
Predict

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Rekik, I., Li, G., Wu, G., Lin, W., & Shen, D. (2015). Prediction of infant MRI appearance and anatomical structure evolution using sparse patch-based metamorphosis learning framework. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9467, pp. 197-204). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9467). Springer Verlag. https://doi.org/10.1007/978-3-319-28194-0_24

Prediction of infant MRI appearance and anatomical structure evolution using sparse patch-based metamorphosis learning framework. / Rekik, Islem; Li, Gang; Wu, Guorong; Lin, Weili; Shen, Dinggang.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 9467 Springer Verlag, 2015. p. 197-204 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9467).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Rekik, I, Li, G, Wu, G, Lin, W & Shen, D 2015, Prediction of infant MRI appearance and anatomical structure evolution using sparse patch-based metamorphosis learning framework. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 9467, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9467, Springer Verlag, pp. 197-204, 1st International Workshop on Patch-Based Techniques in Medical Imaging, Patch-MI 2015, Munich, Germany, 15/10/9. https://doi.org/10.1007/978-3-319-28194-0_24
Rekik I, Li G, Wu G, Lin W, Shen D. Prediction of infant MRI appearance and anatomical structure evolution using sparse patch-based metamorphosis learning framework. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 9467. Springer Verlag. 2015. p. 197-204. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-28194-0_24
Rekik, Islem ; Li, Gang ; Wu, Guorong ; Lin, Weili ; Shen, Dinggang. / Prediction of infant MRI appearance and anatomical structure evolution using sparse patch-based metamorphosis learning framework. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 9467 Springer Verlag, 2015. pp. 197-204 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{16893dfbdca94c669ba896ec33f33970,
title = "Prediction of infant MRI appearance and anatomical structure evolution using sparse patch-based metamorphosis learning framework",
abstract = "Magnetic resonance imaging (MRI) of pediatric brain provides invaluable information for early normal and abnormal brain development. Longitudinal neuroimaging has spanned various research works on examining infant brain development patterns. However, studies on predicting postnatal brain image evolution remain scarce, which is very challenging due to the dynamic tissue contrast change and even inversion in postnatal brains. In this paper, we unprecedentedly propose a dual image intensity and anatomical structure (label) prediction framework that nicely links the geodesic image metamorphosis model with sparse patch-based image representation, thereby defining spatiotemporal metamorphic patches encoding both image photometric and geometric deformation. In the training stage, we learn the 4D metamorphosis trajectories for each training subject. In the prediction stage, we define various strategies to sparsely represent each patch in the testing image using the training metamorphosis patches; while progressively incrementing the richness of the patch (from appearance-based to multimodal kinetic patches). We used the proposed framework to predict 6, 9 and 12-month brain MR image intensity and structure (white and gray matter maps) from 3 months in 10 infants. The proposed framework showed promising preliminary prediction results for the spatiotemporally complex, drastically changing brain images.",
author = "Islem Rekik and Gang Li and Guorong Wu and Weili Lin and Dinggang Shen",
year = "2015",
doi = "10.1007/978-3-319-28194-0_24",
language = "English",
isbn = "9783319281933",
volume = "9467",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "197--204",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - Prediction of infant MRI appearance and anatomical structure evolution using sparse patch-based metamorphosis learning framework

AU - Rekik, Islem

AU - Li, Gang

AU - Wu, Guorong

AU - Lin, Weili

AU - Shen, Dinggang

PY - 2015

Y1 - 2015

N2 - Magnetic resonance imaging (MRI) of pediatric brain provides invaluable information for early normal and abnormal brain development. Longitudinal neuroimaging has spanned various research works on examining infant brain development patterns. However, studies on predicting postnatal brain image evolution remain scarce, which is very challenging due to the dynamic tissue contrast change and even inversion in postnatal brains. In this paper, we unprecedentedly propose a dual image intensity and anatomical structure (label) prediction framework that nicely links the geodesic image metamorphosis model with sparse patch-based image representation, thereby defining spatiotemporal metamorphic patches encoding both image photometric and geometric deformation. In the training stage, we learn the 4D metamorphosis trajectories for each training subject. In the prediction stage, we define various strategies to sparsely represent each patch in the testing image using the training metamorphosis patches; while progressively incrementing the richness of the patch (from appearance-based to multimodal kinetic patches). We used the proposed framework to predict 6, 9 and 12-month brain MR image intensity and structure (white and gray matter maps) from 3 months in 10 infants. The proposed framework showed promising preliminary prediction results for the spatiotemporally complex, drastically changing brain images.

AB - Magnetic resonance imaging (MRI) of pediatric brain provides invaluable information for early normal and abnormal brain development. Longitudinal neuroimaging has spanned various research works on examining infant brain development patterns. However, studies on predicting postnatal brain image evolution remain scarce, which is very challenging due to the dynamic tissue contrast change and even inversion in postnatal brains. In this paper, we unprecedentedly propose a dual image intensity and anatomical structure (label) prediction framework that nicely links the geodesic image metamorphosis model with sparse patch-based image representation, thereby defining spatiotemporal metamorphic patches encoding both image photometric and geometric deformation. In the training stage, we learn the 4D metamorphosis trajectories for each training subject. In the prediction stage, we define various strategies to sparsely represent each patch in the testing image using the training metamorphosis patches; while progressively incrementing the richness of the patch (from appearance-based to multimodal kinetic patches). We used the proposed framework to predict 6, 9 and 12-month brain MR image intensity and structure (white and gray matter maps) from 3 months in 10 infants. The proposed framework showed promising preliminary prediction results for the spatiotemporally complex, drastically changing brain images.

UR - http://www.scopus.com/inward/record.url?scp=84955249351&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84955249351&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-28194-0_24

DO - 10.1007/978-3-319-28194-0_24

M3 - Conference contribution

AN - SCOPUS:84955249351

SN - 9783319281933

VL - 9467

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 197

EP - 204

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

PB - Springer Verlag

ER -