Regression guided deformable models for segmentation of multiple brain ROIs

Zhengwang Wu, Sang Hyun Park, Yanrong Guo, Yaozong Gao, Dinggang Shen

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

1 Citation (Scopus)

Abstract

This paper proposes a novel method of using regressionguided deformable models for brain regions of interest (ROIs) segmentation. Different from conventional deformable segmentation, which often deforms shape model locally and thus sensitive to initialization, we propose to learn a regressor to explicitly guide the shape deformation, thus eventually improves the performance of ROI segmentation. The regressor is learned via two steps, (1) a joint classification and regression random forest (CRRF) and (2) an auto-context model. The CRRF predicts each voxel’s deformation to the nearest point on the ROI boundary as well as each voxel’s class label (e.g., ROI versus background). The auto-context model further refines all voxel’s deformations (i.e., deformation field) and class labels (i.e., label maps) by considering the neighboring structures. Compared to the conventional random forest regressor, the proposed regressor provides more accurate deformation field estimation and thus more robust in guiding deformation of the shape model. Validated in segmentation of 14 midbrain ROIs from the IXI dataset, our method outperforms the state-of-art multi-atlas label fusion and classification methods, and also significantly reduces the computation cost.

Original languageEnglish
Title of host publicationMachine Learning in Medical Imaging - 7th International Workshop, MLMI 2016 held in conjunction with MICCAI 2016, Proceedings
PublisherSpringer Verlag
Pages237-245
Number of pages9
Volume10019 LNCS
ISBN (Print)9783319471563
DOIs
Publication statusPublished - 2016
Externally publishedYes
Event7th International Workshop on Machine Learning in Medical Imaging, MLMI 2016 held in conjunction with 19th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2016 - Athens, Greece
Duration: 2016 Oct 172016 Oct 17

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10019 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other7th International Workshop on Machine Learning in Medical Imaging, MLMI 2016 held in conjunction with 19th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2016
CountryGreece
CityAthens
Period16/10/1716/10/17

Fingerprint

Deformable Models
Region of Interest
Brain
Segmentation
Regression
Random Forest
Labels
Voxel
Atlas
Initialization
Model
Fusion
Fusion reactions
Predict
Costs

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Wu, Z., Park, S. H., Guo, Y., Gao, Y., & Shen, D. (2016). Regression guided deformable models for segmentation of multiple brain ROIs. In Machine Learning in Medical Imaging - 7th International Workshop, MLMI 2016 held in conjunction with MICCAI 2016, Proceedings (Vol. 10019 LNCS, pp. 237-245). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10019 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-47157-0_29

Regression guided deformable models for segmentation of multiple brain ROIs. / Wu, Zhengwang; Park, Sang Hyun; Guo, Yanrong; Gao, Yaozong; Shen, Dinggang.

Machine Learning in Medical Imaging - 7th International Workshop, MLMI 2016 held in conjunction with MICCAI 2016, Proceedings. Vol. 10019 LNCS Springer Verlag, 2016. p. 237-245 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10019 LNCS).

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

Wu, Z, Park, SH, Guo, Y, Gao, Y & Shen, D 2016, Regression guided deformable models for segmentation of multiple brain ROIs. in Machine Learning in Medical Imaging - 7th International Workshop, MLMI 2016 held in conjunction with MICCAI 2016, Proceedings. vol. 10019 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10019 LNCS, Springer Verlag, pp. 237-245, 7th International Workshop on Machine Learning in Medical Imaging, MLMI 2016 held in conjunction with 19th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2016, Athens, Greece, 16/10/17. https://doi.org/10.1007/978-3-319-47157-0_29
Wu Z, Park SH, Guo Y, Gao Y, Shen D. Regression guided deformable models for segmentation of multiple brain ROIs. In Machine Learning in Medical Imaging - 7th International Workshop, MLMI 2016 held in conjunction with MICCAI 2016, Proceedings. Vol. 10019 LNCS. Springer Verlag. 2016. p. 237-245. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-47157-0_29
Wu, Zhengwang ; Park, Sang Hyun ; Guo, Yanrong ; Gao, Yaozong ; Shen, Dinggang. / Regression guided deformable models for segmentation of multiple brain ROIs. Machine Learning in Medical Imaging - 7th International Workshop, MLMI 2016 held in conjunction with MICCAI 2016, Proceedings. Vol. 10019 LNCS Springer Verlag, 2016. pp. 237-245 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{e6a8cc680c92496d85da684744afb993,
title = "Regression guided deformable models for segmentation of multiple brain ROIs",
abstract = "This paper proposes a novel method of using regressionguided deformable models for brain regions of interest (ROIs) segmentation. Different from conventional deformable segmentation, which often deforms shape model locally and thus sensitive to initialization, we propose to learn a regressor to explicitly guide the shape deformation, thus eventually improves the performance of ROI segmentation. The regressor is learned via two steps, (1) a joint classification and regression random forest (CRRF) and (2) an auto-context model. The CRRF predicts each voxel’s deformation to the nearest point on the ROI boundary as well as each voxel’s class label (e.g., ROI versus background). The auto-context model further refines all voxel’s deformations (i.e., deformation field) and class labels (i.e., label maps) by considering the neighboring structures. Compared to the conventional random forest regressor, the proposed regressor provides more accurate deformation field estimation and thus more robust in guiding deformation of the shape model. Validated in segmentation of 14 midbrain ROIs from the IXI dataset, our method outperforms the state-of-art multi-atlas label fusion and classification methods, and also significantly reduces the computation cost.",
author = "Zhengwang Wu and Park, {Sang Hyun} and Yanrong Guo and Yaozong Gao and Dinggang Shen",
year = "2016",
doi = "10.1007/978-3-319-47157-0_29",
language = "English",
isbn = "9783319471563",
volume = "10019 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "237--245",
booktitle = "Machine Learning in Medical Imaging - 7th International Workshop, MLMI 2016 held in conjunction with MICCAI 2016, Proceedings",

}

TY - GEN

T1 - Regression guided deformable models for segmentation of multiple brain ROIs

AU - Wu, Zhengwang

AU - Park, Sang Hyun

AU - Guo, Yanrong

AU - Gao, Yaozong

AU - Shen, Dinggang

PY - 2016

Y1 - 2016

N2 - This paper proposes a novel method of using regressionguided deformable models for brain regions of interest (ROIs) segmentation. Different from conventional deformable segmentation, which often deforms shape model locally and thus sensitive to initialization, we propose to learn a regressor to explicitly guide the shape deformation, thus eventually improves the performance of ROI segmentation. The regressor is learned via two steps, (1) a joint classification and regression random forest (CRRF) and (2) an auto-context model. The CRRF predicts each voxel’s deformation to the nearest point on the ROI boundary as well as each voxel’s class label (e.g., ROI versus background). The auto-context model further refines all voxel’s deformations (i.e., deformation field) and class labels (i.e., label maps) by considering the neighboring structures. Compared to the conventional random forest regressor, the proposed regressor provides more accurate deformation field estimation and thus more robust in guiding deformation of the shape model. Validated in segmentation of 14 midbrain ROIs from the IXI dataset, our method outperforms the state-of-art multi-atlas label fusion and classification methods, and also significantly reduces the computation cost.

AB - This paper proposes a novel method of using regressionguided deformable models for brain regions of interest (ROIs) segmentation. Different from conventional deformable segmentation, which often deforms shape model locally and thus sensitive to initialization, we propose to learn a regressor to explicitly guide the shape deformation, thus eventually improves the performance of ROI segmentation. The regressor is learned via two steps, (1) a joint classification and regression random forest (CRRF) and (2) an auto-context model. The CRRF predicts each voxel’s deformation to the nearest point on the ROI boundary as well as each voxel’s class label (e.g., ROI versus background). The auto-context model further refines all voxel’s deformations (i.e., deformation field) and class labels (i.e., label maps) by considering the neighboring structures. Compared to the conventional random forest regressor, the proposed regressor provides more accurate deformation field estimation and thus more robust in guiding deformation of the shape model. Validated in segmentation of 14 midbrain ROIs from the IXI dataset, our method outperforms the state-of-art multi-atlas label fusion and classification methods, and also significantly reduces the computation cost.

UR - http://www.scopus.com/inward/record.url?scp=84992478806&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84992478806&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-47157-0_29

DO - 10.1007/978-3-319-47157-0_29

M3 - Conference contribution

AN - SCOPUS:84992478806

SN - 9783319471563

VL - 10019 LNCS

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

SP - 237

EP - 245

BT - Machine Learning in Medical Imaging - 7th International Workshop, MLMI 2016 held in conjunction with MICCAI 2016, Proceedings

PB - Springer Verlag

ER -