Self-training for Brain Tumour Segmentation with Uncertainty Estimation and Biophysics-Guided Survival Prediction

Chengliang Dai; Shuo Wang; Hadrien Raynaud; Yuanhan Mo; Elsa Angelini; Yike Guo; Wenjia Bai

doi:10.1007/978-3-030-72084-1_46

Self-training for Brain Tumour Segmentation with Uncertainty Estimation and Biophysics-Guided Survival Prediction

Chengliang Dai^*, Shuo Wang, Hadrien Raynaud, Yuanhan Mo, Elsa Angelini, Yike Guo, Wenjia Bai

^*Corresponding author for this work

Research output: Chapter in Book/Conference Proceeding/Report › Conference Paper published in a book › peer-review

4 Citations (Scopus)

Abstract

Gliomas are among the most common types of malignant brain tumours in adults. Given the intrinsic heterogeneity of gliomas, the multi-parametric magnetic resonance imaging (mpMRI) is the most effective technique for characterising gliomas and their sub-regions. Accurate segmentation of the tumour sub-regions on mpMRI is of clinical significance, which provides valuable information for treatment planning and survival prediction. Thanks to the recent developments on deep learning, the accuracy of automated medical image segmentation has improved significantly. In this paper, we leverage the widely used attention and self-training techniques to conduct reliable brain tumour segmentation and uncertainty estimation. Based on the segmentation result, we present a biophysics-guided prognostic model for the prediction of overall survival. Our method of uncertainty estimation has won the second place of the MICCAI 2020 BraTS Challenge.

Original language	English
Title of host publication	Brainlesion
Subtitle of host publication	Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - 6th International Workshop, BrainLes 2020, Held in Conjunction with MICCAI 2020, Revised Selected Papers
Editors	Alessandro Crimi, Spyridon Bakas
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	514-523
Number of pages	10
ISBN (Print)	9783030720834
DOIs	https://doi.org/10.1007/978-3-030-72084-1_46
Publication status	Published - 2021
Externally published	Yes
Event	6th International MICCAI Brainlesion Workshop, BrainLes 2020 Held in Conjunction with 23rd Medical Image Computing for Computer Assisted Intervention Conference, MICCAI 2020 - Virtual, Online Duration: 4 Oct 2020 → 4 Oct 2020

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	12658 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	6th International MICCAI Brainlesion Workshop, BrainLes 2020 Held in Conjunction with 23rd Medical Image Computing for Computer Assisted Intervention Conference, MICCAI 2020
City	Virtual, Online
Period	4/10/20 → 4/10/20

Bibliographical note

Publisher Copyright:
© 2021, Springer Nature Switzerland AG.

Keywords

Brain imaging
Deep learning
Radiomics
Tumour segmentation

Access to Document

10.1007/978-3-030-72084-1_46

Cite this

Dai, C., Wang, S., Raynaud, H., Mo, Y., Angelini, E., Guo, Y., & Bai, W. (2021). Self-training for Brain Tumour Segmentation with Uncertainty Estimation and Biophysics-Guided Survival Prediction. In A. Crimi, & S. Bakas (Eds.), Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - 6th International Workshop, BrainLes 2020, Held in Conjunction with MICCAI 2020, Revised Selected Papers (pp. 514-523). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12658 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-72084-1_46

Dai, Chengliang ; Wang, Shuo ; Raynaud, Hadrien et al. / Self-training for Brain Tumour Segmentation with Uncertainty Estimation and Biophysics-Guided Survival Prediction. Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - 6th International Workshop, BrainLes 2020, Held in Conjunction with MICCAI 2020, Revised Selected Papers. editor / Alessandro Crimi ; Spyridon Bakas. Springer Science and Business Media Deutschland GmbH, 2021. pp. 514-523 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{ad139f7a3f1745f68d00754d0b5564a0,

title = "Self-training for Brain Tumour Segmentation with Uncertainty Estimation and Biophysics-Guided Survival Prediction",

abstract = "Gliomas are among the most common types of malignant brain tumours in adults. Given the intrinsic heterogeneity of gliomas, the multi-parametric magnetic resonance imaging (mpMRI) is the most effective technique for characterising gliomas and their sub-regions. Accurate segmentation of the tumour sub-regions on mpMRI is of clinical significance, which provides valuable information for treatment planning and survival prediction. Thanks to the recent developments on deep learning, the accuracy of automated medical image segmentation has improved significantly. In this paper, we leverage the widely used attention and self-training techniques to conduct reliable brain tumour segmentation and uncertainty estimation. Based on the segmentation result, we present a biophysics-guided prognostic model for the prediction of overall survival. Our method of uncertainty estimation has won the second place of the MICCAI 2020 BraTS Challenge.",

keywords = "Brain imaging, Deep learning, Radiomics, Tumour segmentation",

author = "Chengliang Dai and Shuo Wang and Hadrien Raynaud and Yuanhan Mo and Elsa Angelini and Yike Guo and Wenjia Bai",

note = "Publisher Copyright: {\textcopyright} 2021, Springer Nature Switzerland AG.; 6th International MICCAI Brainlesion Workshop, BrainLes 2020 Held in Conjunction with 23rd Medical Image Computing for Computer Assisted Intervention Conference, MICCAI 2020 ; Conference date: 04-10-2020 Through 04-10-2020",

year = "2021",

doi = "10.1007/978-3-030-72084-1\_46",

language = "English",

isbn = "9783030720834",

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

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "514--523",

editor = "Alessandro Crimi and Spyridon Bakas",

booktitle = "Brainlesion",

}

Dai, C, Wang, S, Raynaud, H, Mo, Y, Angelini, E, Guo, Y & Bai, W 2021, Self-training for Brain Tumour Segmentation with Uncertainty Estimation and Biophysics-Guided Survival Prediction. in A Crimi & S Bakas (eds), Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - 6th International Workshop, BrainLes 2020, Held in Conjunction with MICCAI 2020, Revised Selected Papers. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 12658 LNCS, Springer Science and Business Media Deutschland GmbH, pp. 514-523, 6th International MICCAI Brainlesion Workshop, BrainLes 2020 Held in Conjunction with 23rd Medical Image Computing for Computer Assisted Intervention Conference, MICCAI 2020, Virtual, Online, 4/10/20. https://doi.org/10.1007/978-3-030-72084-1_46

Self-training for Brain Tumour Segmentation with Uncertainty Estimation and Biophysics-Guided Survival Prediction. / Dai, Chengliang; Wang, Shuo; Raynaud, Hadrien et al.
Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - 6th International Workshop, BrainLes 2020, Held in Conjunction with MICCAI 2020, Revised Selected Papers. ed. / Alessandro Crimi; Spyridon Bakas. Springer Science and Business Media Deutschland GmbH, 2021. p. 514-523 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12658 LNCS).

Research output: Chapter in Book/Conference Proceeding/Report › Conference Paper published in a book › peer-review

TY - GEN

T1 - Self-training for Brain Tumour Segmentation with Uncertainty Estimation and Biophysics-Guided Survival Prediction

AU - Dai, Chengliang

AU - Wang, Shuo

AU - Raynaud, Hadrien

AU - Mo, Yuanhan

AU - Angelini, Elsa

AU - Guo, Yike

AU - Bai, Wenjia

PY - 2021

Y1 - 2021

N2 - Gliomas are among the most common types of malignant brain tumours in adults. Given the intrinsic heterogeneity of gliomas, the multi-parametric magnetic resonance imaging (mpMRI) is the most effective technique for characterising gliomas and their sub-regions. Accurate segmentation of the tumour sub-regions on mpMRI is of clinical significance, which provides valuable information for treatment planning and survival prediction. Thanks to the recent developments on deep learning, the accuracy of automated medical image segmentation has improved significantly. In this paper, we leverage the widely used attention and self-training techniques to conduct reliable brain tumour segmentation and uncertainty estimation. Based on the segmentation result, we present a biophysics-guided prognostic model for the prediction of overall survival. Our method of uncertainty estimation has won the second place of the MICCAI 2020 BraTS Challenge.

AB - Gliomas are among the most common types of malignant brain tumours in adults. Given the intrinsic heterogeneity of gliomas, the multi-parametric magnetic resonance imaging (mpMRI) is the most effective technique for characterising gliomas and their sub-regions. Accurate segmentation of the tumour sub-regions on mpMRI is of clinical significance, which provides valuable information for treatment planning and survival prediction. Thanks to the recent developments on deep learning, the accuracy of automated medical image segmentation has improved significantly. In this paper, we leverage the widely used attention and self-training techniques to conduct reliable brain tumour segmentation and uncertainty estimation. Based on the segmentation result, we present a biophysics-guided prognostic model for the prediction of overall survival. Our method of uncertainty estimation has won the second place of the MICCAI 2020 BraTS Challenge.

KW - Brain imaging

KW - Deep learning

KW - Radiomics

KW - Tumour segmentation

UR - https://www.webofscience.com/wos/woscc/full-record/WOS:000892566900046

UR - https://openalex.org/W3149842999

UR - https://www.scopus.com/pages/publications/85107353117

U2 - 10.1007/978-3-030-72084-1_46

DO - 10.1007/978-3-030-72084-1_46

M3 - Conference Paper published in a book

SN - 9783030720834

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

SP - 514

EP - 523

BT - Brainlesion

A2 - Crimi, Alessandro

A2 - Bakas, Spyridon

PB - Springer Science and Business Media Deutschland GmbH

T2 - 6th International MICCAI Brainlesion Workshop, BrainLes 2020 Held in Conjunction with 23rd Medical Image Computing for Computer Assisted Intervention Conference, MICCAI 2020

Y2 - 4 October 2020 through 4 October 2020

ER -

Dai C, Wang S, Raynaud H, Mo Y, Angelini E, Guo Y et al. Self-training for Brain Tumour Segmentation with Uncertainty Estimation and Biophysics-Guided Survival Prediction. In Crimi A, Bakas S, editors, Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - 6th International Workshop, BrainLes 2020, Held in Conjunction with MICCAI 2020, Revised Selected Papers. Springer Science and Business Media Deutschland GmbH. 2021. p. 514-523. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-72084-1_46

Self-training for Brain Tumour Segmentation with Uncertainty Estimation and Biophysics-Guided Survival Prediction

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this