Lifted auto-context forests for brain tumour segmentation

Loic Le Folgoc; Aditya V. Nori; Siddharth Ancha; Antonio Criminisi

doi:10.1007/978-3-319-55524-9_17

Lifted auto-context forests for brain tumour segmentation

Loic Le Folgoc, Aditya V. Nori, Siddharth Ancha, Antonio Criminisi

Microsoft Research Cambridge

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

We revisit Auto-Context Forests for brain tumour segmentation in multi-channel magnetic resonance images, where semantic context is progressively built and refined via successive layers of Decision Forests (DFs). Specifically, we make the following contributions: (1) improved generalization via an efficient node-splitting criterion based on holdout estimates, (2) increased compactness at a tree-level, thereby yielding shallow discriminative ensembles trained orders of magnitude faster, and (3) guided semantic bagging that exposes latent data-space semantics captured by forest pathways. The proposed framework is practical: the per-layer training is fast, modular and robust. It was a top performer in the MICCAI 2016 BRATS (Brain Tumour Segmentation) challenge, and this paper aims to discuss and provide details about the challenge entry.

Original language	English
Title of host publication	Brainlesion
Subtitle of host publication	Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - Second International Workshop, BrainLes 2016, with the Challenges on BRATS, ISLES and mTOP 2016 Held in Conjunction with MICCAI 2016, Revised Selected Papers
Editors	Bjoern Menze, Mauricio Reyes, Alessandro Crimi, Oskar Maier, Stefan Winzeck, Heinz Handels
Publisher	Springer Verlag
Pages	171-183
Number of pages	13
ISBN (Print)	9783319555232
DOIs	https://doi.org/10.1007/978-3-319-55524-9_17
Publication status	Published - 1 Jan 2016
Externally published	Yes
Event	2nd International Workshop on Brain Lesion, BrainLes 2016, with the challenges on Brain Tumor Segmentation BRATS, Ischemic Stroke Lesion Image Segmentation ISLES, and the Mild Traumatic Brain Injury Outcome Prediction mTOP held in conjunction with the International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2016 - Athens, Greece Duration: 17 Oct 2016 → 17 Oct 2016

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	10154 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	2nd International Workshop on Brain Lesion, BrainLes 2016, with the challenges on Brain Tumor Segmentation BRATS, Ischemic Stroke Lesion Image Segmentation ISLES, and the Mild Traumatic Brain Injury Outcome Prediction mTOP held in conjunction with the International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2016
Country/Territory	Greece
City	Athens
Period	17/10/16 → 17/10/16

Access to Document

10.1007/978-3-319-55524-9_17

Cite this

Le Folgoc, L., Nori, A. V., Ancha, S., & Criminisi, A. (2016). Lifted auto-context forests for brain tumour segmentation. In B. Menze, M. Reyes, A. Crimi, O. Maier, S. Winzeck, & H. Handels (Eds.), Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - Second International Workshop, BrainLes 2016, with the Challenges on BRATS, ISLES and mTOP 2016 Held in Conjunction with MICCAI 2016, Revised Selected Papers (pp. 171-183). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10154 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-55524-9_17

Le Folgoc, Loic ; Nori, Aditya V. ; Ancha, Siddharth et al. / Lifted auto-context forests for brain tumour segmentation. Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - Second International Workshop, BrainLes 2016, with the Challenges on BRATS, ISLES and mTOP 2016 Held in Conjunction with MICCAI 2016, Revised Selected Papers. editor / Bjoern Menze ; Mauricio Reyes ; Alessandro Crimi ; Oskar Maier ; Stefan Winzeck ; Heinz Handels. Springer Verlag, 2016. pp. 171-183 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{43086762a5254b599a49b726b6a94acc,

title = "Lifted auto-context forests for brain tumour segmentation",

abstract = "We revisit Auto-Context Forests for brain tumour segmentation in multi-channel magnetic resonance images, where semantic context is progressively built and refined via successive layers of Decision Forests (DFs). Specifically, we make the following contributions: (1) improved generalization via an efficient node-splitting criterion based on holdout estimates, (2) increased compactness at a tree-level, thereby yielding shallow discriminative ensembles trained orders of magnitude faster, and (3) guided semantic bagging that exposes latent data-space semantics captured by forest pathways. The proposed framework is practical: the per-layer training is fast, modular and robust. It was a top performer in the MICCAI 2016 BRATS (Brain Tumour Segmentation) challenge, and this paper aims to discuss and provide details about the challenge entry.",

author = "\{Le Folgoc\}, Loic and Nori, \{Aditya V.\} and Siddharth Ancha and Antonio Criminisi",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing AG 2016.; 2nd International Workshop on Brain Lesion, BrainLes 2016, with the challenges on Brain Tumor Segmentation BRATS, Ischemic Stroke Lesion Image Segmentation ISLES, and the Mild Traumatic Brain Injury Outcome Prediction mTOP held in conjunction with the International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2016 ; Conference date: 17-10-2016 Through 17-10-2016",

year = "2016",

month = jan,

day = "1",

doi = "10.1007/978-3-319-55524-9\_17",

language = "English",

isbn = "9783319555232",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "171--183",

editor = "Bjoern Menze and Mauricio Reyes and Alessandro Crimi and Oskar Maier and Stefan Winzeck and Heinz Handels",

booktitle = "Brainlesion",

}

Le Folgoc, L, Nori, AV, Ancha, S & Criminisi, A 2016, Lifted auto-context forests for brain tumour segmentation. in B Menze, M Reyes, A Crimi, O Maier, S Winzeck & H Handels (eds), Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - Second International Workshop, BrainLes 2016, with the Challenges on BRATS, ISLES and mTOP 2016 Held in Conjunction with MICCAI 2016, Revised Selected Papers. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10154 LNCS, Springer Verlag, pp. 171-183, 2nd International Workshop on Brain Lesion, BrainLes 2016, with the challenges on Brain Tumor Segmentation BRATS, Ischemic Stroke Lesion Image Segmentation ISLES, and the Mild Traumatic Brain Injury Outcome Prediction mTOP held in conjunction with the International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2016, Athens, Greece, 17/10/16. https://doi.org/10.1007/978-3-319-55524-9_17

Lifted auto-context forests for brain tumour segmentation. / Le Folgoc, Loic; Nori, Aditya V.; Ancha, Siddharth et al.
Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - Second International Workshop, BrainLes 2016, with the Challenges on BRATS, ISLES and mTOP 2016 Held in Conjunction with MICCAI 2016, Revised Selected Papers. ed. / Bjoern Menze; Mauricio Reyes; Alessandro Crimi; Oskar Maier; Stefan Winzeck; Heinz Handels. Springer Verlag, 2016. p. 171-183 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10154 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Lifted auto-context forests for brain tumour segmentation

AU - Le Folgoc, Loic

AU - Nori, Aditya V.

AU - Ancha, Siddharth

AU - Criminisi, Antonio

N1 - Publisher Copyright: © Springer International Publishing AG 2016.

PY - 2016/1/1

Y1 - 2016/1/1

N2 - We revisit Auto-Context Forests for brain tumour segmentation in multi-channel magnetic resonance images, where semantic context is progressively built and refined via successive layers of Decision Forests (DFs). Specifically, we make the following contributions: (1) improved generalization via an efficient node-splitting criterion based on holdout estimates, (2) increased compactness at a tree-level, thereby yielding shallow discriminative ensembles trained orders of magnitude faster, and (3) guided semantic bagging that exposes latent data-space semantics captured by forest pathways. The proposed framework is practical: the per-layer training is fast, modular and robust. It was a top performer in the MICCAI 2016 BRATS (Brain Tumour Segmentation) challenge, and this paper aims to discuss and provide details about the challenge entry.

AB - We revisit Auto-Context Forests for brain tumour segmentation in multi-channel magnetic resonance images, where semantic context is progressively built and refined via successive layers of Decision Forests (DFs). Specifically, we make the following contributions: (1) improved generalization via an efficient node-splitting criterion based on holdout estimates, (2) increased compactness at a tree-level, thereby yielding shallow discriminative ensembles trained orders of magnitude faster, and (3) guided semantic bagging that exposes latent data-space semantics captured by forest pathways. The proposed framework is practical: the per-layer training is fast, modular and robust. It was a top performer in the MICCAI 2016 BRATS (Brain Tumour Segmentation) challenge, and this paper aims to discuss and provide details about the challenge entry.

U2 - 10.1007/978-3-319-55524-9_17

DO - 10.1007/978-3-319-55524-9_17

M3 - Conference contribution

AN - SCOPUS:85018713475

SN - 9783319555232

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

SP - 171

EP - 183

BT - Brainlesion

A2 - Menze, Bjoern

A2 - Reyes, Mauricio

A2 - Crimi, Alessandro

A2 - Maier, Oskar

A2 - Winzeck, Stefan

A2 - Handels, Heinz

PB - Springer Verlag

T2 - 2nd International Workshop on Brain Lesion, BrainLes 2016, with the challenges on Brain Tumor Segmentation BRATS, Ischemic Stroke Lesion Image Segmentation ISLES, and the Mild Traumatic Brain Injury Outcome Prediction mTOP held in conjunction with the International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2016

Y2 - 17 October 2016 through 17 October 2016

ER -

Le Folgoc L, Nori AV, Ancha S, Criminisi A. Lifted auto-context forests for brain tumour segmentation. In Menze B, Reyes M, Crimi A, Maier O, Winzeck S, Handels H, editors, Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - Second International Workshop, BrainLes 2016, with the Challenges on BRATS, ISLES and mTOP 2016 Held in Conjunction with MICCAI 2016, Revised Selected Papers. Springer Verlag. 2016. p. 171-183. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-55524-9_17

Lifted auto-context forests for brain tumour segmentation

Abstract

Publication series

Conference

Access to Document

Fingerprint

Cite this