Labeling of the cerebellar peduncles using a supervised Gaussian classifier with volumetric tract segmentation

Chuyang Ye, Pierre Louis Bazin, John A. Bogovic, Sarah H. Ying, Jerry Ladd Prince

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

Abstract

The cerebellar peduncles are white matter tracts that play an important role in the communication of the cerebellum with other regions of the brain. They can be grouped into three fiber bundles: inferior cerebellar peduncle, middle cerebellar peduncle, and superior cerebellar peduncle. Their automatic segmentation on diffusion tensor images would enable a better understanding of the cerebellum and would be less time-consuming and more reproducible than manual delineation. This paper presents a method that automatically labels the three fiber bundles based on the segmentation results from the diffusion oriented tract segmentation (DOTS) algorithm, which achieves volume segmentation of white matter tracts using a Markov random field (MRF) framework. We use the DOTS labeling result as a guide to determine the classification of fibers produced by wild bootstrap probabilistic tractography. Mean distances from each fiber to each DOTS volume label are defined and then used as features that contribute to classification. A supervised Gaussian classifier is employed to label the fibers. Manually delineated cerebellar peduncles serve as training data to determine the parameters of class probabilities for each label. Fibers are labeled as the class that has the highest posterior probability. An outlier detection step removes fiber tracts that belong to noise or that are not modeled by DOTS. Experiments show a successful classification of the cerebellar peduncles. We have also compared results between successive scans to demonstrate the reproducibility of the proposed method.

Original languageEnglish (US)
Title of host publicationProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume8314
DOIs
StatePublished - 2012
EventMedical Imaging 2012: Image Processing - San Diego, CA, United States
Duration: Feb 6 2012Feb 9 2012

Other

OtherMedical Imaging 2012: Image Processing
CountryUnited States
CitySan Diego, CA
Period2/6/122/9/12

Fingerprint

classifiers
Labeling
marking
Classifiers
fibers
Fibers
Labels
cerebellum
Cerebellum
bundles
Noise
delineation
brain
Tensors
Brain
education
communication
tensors
Communication
White Matter

Keywords

  • Cerebellar peduncles
  • DOTS
  • DTI
  • Fiber labeling
  • Supervised Gaussian classifier

ASJC Scopus subject areas

  • Atomic and Molecular Physics, and Optics
  • Electronic, Optical and Magnetic Materials
  • Biomaterials
  • Radiology Nuclear Medicine and imaging

Cite this

Ye, C., Bazin, P. L., Bogovic, J. A., Ying, S. H., & Prince, J. L. (2012). Labeling of the cerebellar peduncles using a supervised Gaussian classifier with volumetric tract segmentation. In Progress in Biomedical Optics and Imaging - Proceedings of SPIE (Vol. 8314). [831447] https://doi.org/10.1117/12.910551

Labeling of the cerebellar peduncles using a supervised Gaussian classifier with volumetric tract segmentation. / Ye, Chuyang; Bazin, Pierre Louis; Bogovic, John A.; Ying, Sarah H.; Prince, Jerry Ladd.

Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 8314 2012. 831447.

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

Ye, C, Bazin, PL, Bogovic, JA, Ying, SH & Prince, JL 2012, Labeling of the cerebellar peduncles using a supervised Gaussian classifier with volumetric tract segmentation. in Progress in Biomedical Optics and Imaging - Proceedings of SPIE. vol. 8314, 831447, Medical Imaging 2012: Image Processing, San Diego, CA, United States, 2/6/12. https://doi.org/10.1117/12.910551
Ye C, Bazin PL, Bogovic JA, Ying SH, Prince JL. Labeling of the cerebellar peduncles using a supervised Gaussian classifier with volumetric tract segmentation. In Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 8314. 2012. 831447 https://doi.org/10.1117/12.910551
Ye, Chuyang ; Bazin, Pierre Louis ; Bogovic, John A. ; Ying, Sarah H. ; Prince, Jerry Ladd. / Labeling of the cerebellar peduncles using a supervised Gaussian classifier with volumetric tract segmentation. Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 8314 2012.
@inproceedings{2247630811e84b93adac0cc1cb460728,
title = "Labeling of the cerebellar peduncles using a supervised Gaussian classifier with volumetric tract segmentation",
abstract = "The cerebellar peduncles are white matter tracts that play an important role in the communication of the cerebellum with other regions of the brain. They can be grouped into three fiber bundles: inferior cerebellar peduncle, middle cerebellar peduncle, and superior cerebellar peduncle. Their automatic segmentation on diffusion tensor images would enable a better understanding of the cerebellum and would be less time-consuming and more reproducible than manual delineation. This paper presents a method that automatically labels the three fiber bundles based on the segmentation results from the diffusion oriented tract segmentation (DOTS) algorithm, which achieves volume segmentation of white matter tracts using a Markov random field (MRF) framework. We use the DOTS labeling result as a guide to determine the classification of fibers produced by wild bootstrap probabilistic tractography. Mean distances from each fiber to each DOTS volume label are defined and then used as features that contribute to classification. A supervised Gaussian classifier is employed to label the fibers. Manually delineated cerebellar peduncles serve as training data to determine the parameters of class probabilities for each label. Fibers are labeled as the class that has the highest posterior probability. An outlier detection step removes fiber tracts that belong to noise or that are not modeled by DOTS. Experiments show a successful classification of the cerebellar peduncles. We have also compared results between successive scans to demonstrate the reproducibility of the proposed method.",
keywords = "Cerebellar peduncles, DOTS, DTI, Fiber labeling, Supervised Gaussian classifier",
author = "Chuyang Ye and Bazin, {Pierre Louis} and Bogovic, {John A.} and Ying, {Sarah H.} and Prince, {Jerry Ladd}",
year = "2012",
doi = "10.1117/12.910551",
language = "English (US)",
isbn = "9780819489630",
volume = "8314",
booktitle = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",

}

TY - GEN

T1 - Labeling of the cerebellar peduncles using a supervised Gaussian classifier with volumetric tract segmentation

AU - Ye, Chuyang

AU - Bazin, Pierre Louis

AU - Bogovic, John A.

AU - Ying, Sarah H.

AU - Prince, Jerry Ladd

PY - 2012

Y1 - 2012

N2 - The cerebellar peduncles are white matter tracts that play an important role in the communication of the cerebellum with other regions of the brain. They can be grouped into three fiber bundles: inferior cerebellar peduncle, middle cerebellar peduncle, and superior cerebellar peduncle. Their automatic segmentation on diffusion tensor images would enable a better understanding of the cerebellum and would be less time-consuming and more reproducible than manual delineation. This paper presents a method that automatically labels the three fiber bundles based on the segmentation results from the diffusion oriented tract segmentation (DOTS) algorithm, which achieves volume segmentation of white matter tracts using a Markov random field (MRF) framework. We use the DOTS labeling result as a guide to determine the classification of fibers produced by wild bootstrap probabilistic tractography. Mean distances from each fiber to each DOTS volume label are defined and then used as features that contribute to classification. A supervised Gaussian classifier is employed to label the fibers. Manually delineated cerebellar peduncles serve as training data to determine the parameters of class probabilities for each label. Fibers are labeled as the class that has the highest posterior probability. An outlier detection step removes fiber tracts that belong to noise or that are not modeled by DOTS. Experiments show a successful classification of the cerebellar peduncles. We have also compared results between successive scans to demonstrate the reproducibility of the proposed method.

AB - The cerebellar peduncles are white matter tracts that play an important role in the communication of the cerebellum with other regions of the brain. They can be grouped into three fiber bundles: inferior cerebellar peduncle, middle cerebellar peduncle, and superior cerebellar peduncle. Their automatic segmentation on diffusion tensor images would enable a better understanding of the cerebellum and would be less time-consuming and more reproducible than manual delineation. This paper presents a method that automatically labels the three fiber bundles based on the segmentation results from the diffusion oriented tract segmentation (DOTS) algorithm, which achieves volume segmentation of white matter tracts using a Markov random field (MRF) framework. We use the DOTS labeling result as a guide to determine the classification of fibers produced by wild bootstrap probabilistic tractography. Mean distances from each fiber to each DOTS volume label are defined and then used as features that contribute to classification. A supervised Gaussian classifier is employed to label the fibers. Manually delineated cerebellar peduncles serve as training data to determine the parameters of class probabilities for each label. Fibers are labeled as the class that has the highest posterior probability. An outlier detection step removes fiber tracts that belong to noise or that are not modeled by DOTS. Experiments show a successful classification of the cerebellar peduncles. We have also compared results between successive scans to demonstrate the reproducibility of the proposed method.

KW - Cerebellar peduncles

KW - DOTS

KW - DTI

KW - Fiber labeling

KW - Supervised Gaussian classifier

UR - http://www.scopus.com/inward/record.url?scp=84860778183&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84860778183&partnerID=8YFLogxK

U2 - 10.1117/12.910551

DO - 10.1117/12.910551

M3 - Conference contribution

AN - SCOPUS:84860778183

SN - 9780819489630

VL - 8314

BT - Progress in Biomedical Optics and Imaging - Proceedings of SPIE

ER -