Thalamic parcellation from multi-modal data using random forest learning

Joshua V. Stough, Chuyang Ye, Sarah H. Ying, Jerry Ladd Prince

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

Abstract

The thalamus sub-cortical gray matter structure consists of contiguous nuclei, each individually responsible for communication between various cerebral cortex and midbrain regions. These nuclei are differentially affected in neurodegenerative diseases such as multiple sclerosis and Alzheimer's. However thalamic parcellation of the nuclei, manual or automatic, is difficult given the limited contrast in any particular magnetic resonance (MR) modality. Several groups have had qualitative success differentiating nuclei based on spatial location and fiber orientation information in diffusion tensor imaging (DTI). In this paper, we extend these principles by combining these discriminating dimensions with structural MR and derived information, and by building random forest learners on the resultant multi-modal features. In training, we form a multi-dimensional feature per voxel, which we associate with a nucleus classification from a manual rater. Learners are trained to differentiate thalamus from background and thalamic nuclei from other nuclei. These learners inform the external forces of a multiple object level set model. Our cross-validated quantitative results on a set of twenty subjects show the efficacy and reproducibility of our results.

Original languageEnglish (US)
Title of host publicationProceedings - International Symposium on Biomedical Imaging
Pages852-855
Number of pages4
DOIs
StatePublished - 2013
Event2013 IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2013 - San Francisco, CA, United States
Duration: Apr 7 2013Apr 11 2013

Other

Other2013 IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2013
CountryUnited States
CitySan Francisco, CA
Period4/7/134/11/13

Fingerprint

Thalamic Nuclei
Magnetic resonance
Thalamus
Magnetic Resonance Spectroscopy
Learning
Diffusion tensor imaging
Neurodegenerative diseases
Diffusion Tensor Imaging
Fiber reinforced materials
Mesencephalon
Reproducibility of Results
Neurodegenerative Diseases
Cerebral Cortex
Multiple Sclerosis
Communication
Forests
Gray Matter

Keywords

  • deformable models
  • Diffusion tensor imaging
  • machine learning
  • object segmentation
  • random forests

ASJC Scopus subject areas

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging

Cite this

Stough, J. V., Ye, C., Ying, S. H., & Prince, J. L. (2013). Thalamic parcellation from multi-modal data using random forest learning. In Proceedings - International Symposium on Biomedical Imaging (pp. 852-855). [6556609] https://doi.org/10.1109/ISBI.2013.6556609

Thalamic parcellation from multi-modal data using random forest learning. / Stough, Joshua V.; Ye, Chuyang; Ying, Sarah H.; Prince, Jerry Ladd.

Proceedings - International Symposium on Biomedical Imaging. 2013. p. 852-855 6556609.

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

Stough, JV, Ye, C, Ying, SH & Prince, JL 2013, Thalamic parcellation from multi-modal data using random forest learning. in Proceedings - International Symposium on Biomedical Imaging., 6556609, pp. 852-855, 2013 IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2013, San Francisco, CA, United States, 4/7/13. https://doi.org/10.1109/ISBI.2013.6556609
Stough JV, Ye C, Ying SH, Prince JL. Thalamic parcellation from multi-modal data using random forest learning. In Proceedings - International Symposium on Biomedical Imaging. 2013. p. 852-855. 6556609 https://doi.org/10.1109/ISBI.2013.6556609
Stough, Joshua V. ; Ye, Chuyang ; Ying, Sarah H. ; Prince, Jerry Ladd. / Thalamic parcellation from multi-modal data using random forest learning. Proceedings - International Symposium on Biomedical Imaging. 2013. pp. 852-855
@inproceedings{781a5f4fe3a44d1aa006b297338076f4,
title = "Thalamic parcellation from multi-modal data using random forest learning",
abstract = "The thalamus sub-cortical gray matter structure consists of contiguous nuclei, each individually responsible for communication between various cerebral cortex and midbrain regions. These nuclei are differentially affected in neurodegenerative diseases such as multiple sclerosis and Alzheimer's. However thalamic parcellation of the nuclei, manual or automatic, is difficult given the limited contrast in any particular magnetic resonance (MR) modality. Several groups have had qualitative success differentiating nuclei based on spatial location and fiber orientation information in diffusion tensor imaging (DTI). In this paper, we extend these principles by combining these discriminating dimensions with structural MR and derived information, and by building random forest learners on the resultant multi-modal features. In training, we form a multi-dimensional feature per voxel, which we associate with a nucleus classification from a manual rater. Learners are trained to differentiate thalamus from background and thalamic nuclei from other nuclei. These learners inform the external forces of a multiple object level set model. Our cross-validated quantitative results on a set of twenty subjects show the efficacy and reproducibility of our results.",
keywords = "deformable models, Diffusion tensor imaging, machine learning, object segmentation, random forests",
author = "Stough, {Joshua V.} and Chuyang Ye and Ying, {Sarah H.} and Prince, {Jerry Ladd}",
year = "2013",
doi = "10.1109/ISBI.2013.6556609",
language = "English (US)",
isbn = "9781467364546",
pages = "852--855",
booktitle = "Proceedings - International Symposium on Biomedical Imaging",

}

TY - GEN

T1 - Thalamic parcellation from multi-modal data using random forest learning

AU - Stough, Joshua V.

AU - Ye, Chuyang

AU - Ying, Sarah H.

AU - Prince, Jerry Ladd

PY - 2013

Y1 - 2013

N2 - The thalamus sub-cortical gray matter structure consists of contiguous nuclei, each individually responsible for communication between various cerebral cortex and midbrain regions. These nuclei are differentially affected in neurodegenerative diseases such as multiple sclerosis and Alzheimer's. However thalamic parcellation of the nuclei, manual or automatic, is difficult given the limited contrast in any particular magnetic resonance (MR) modality. Several groups have had qualitative success differentiating nuclei based on spatial location and fiber orientation information in diffusion tensor imaging (DTI). In this paper, we extend these principles by combining these discriminating dimensions with structural MR and derived information, and by building random forest learners on the resultant multi-modal features. In training, we form a multi-dimensional feature per voxel, which we associate with a nucleus classification from a manual rater. Learners are trained to differentiate thalamus from background and thalamic nuclei from other nuclei. These learners inform the external forces of a multiple object level set model. Our cross-validated quantitative results on a set of twenty subjects show the efficacy and reproducibility of our results.

AB - The thalamus sub-cortical gray matter structure consists of contiguous nuclei, each individually responsible for communication between various cerebral cortex and midbrain regions. These nuclei are differentially affected in neurodegenerative diseases such as multiple sclerosis and Alzheimer's. However thalamic parcellation of the nuclei, manual or automatic, is difficult given the limited contrast in any particular magnetic resonance (MR) modality. Several groups have had qualitative success differentiating nuclei based on spatial location and fiber orientation information in diffusion tensor imaging (DTI). In this paper, we extend these principles by combining these discriminating dimensions with structural MR and derived information, and by building random forest learners on the resultant multi-modal features. In training, we form a multi-dimensional feature per voxel, which we associate with a nucleus classification from a manual rater. Learners are trained to differentiate thalamus from background and thalamic nuclei from other nuclei. These learners inform the external forces of a multiple object level set model. Our cross-validated quantitative results on a set of twenty subjects show the efficacy and reproducibility of our results.

KW - deformable models

KW - Diffusion tensor imaging

KW - machine learning

KW - object segmentation

KW - random forests

UR - http://www.scopus.com/inward/record.url?scp=84881653911&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84881653911&partnerID=8YFLogxK

U2 - 10.1109/ISBI.2013.6556609

DO - 10.1109/ISBI.2013.6556609

M3 - Conference contribution

C2 - 24145869

AN - SCOPUS:84881653911

SN - 9781467364546

SP - 852

EP - 855

BT - Proceedings - International Symposium on Biomedical Imaging

ER -