G-CNN: Adaptive Geometric Convolutional Neural Networks for MRI-Based Skull Stripping

Yifan Li; Chao Li; Yiran Wei; Stephen Price; Carola Bibiane Schönlieb; Xi Chen

doi:10.1007/978-3-031-45087-7_3

G-CNN: Adaptive Geometric Convolutional Neural Networks for MRI-Based Skull Stripping

Yifan Li
, Chao Li
, Yiran Wei
, Stephen Price
, Carola Bibiane Schönlieb
, Xi Chen (Lead / Corresponding author)

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

Abstract

Skull stripping in MRI-based brain imaging involves extraction of brain regions from raw images. While some Convolutional Neural Nets (CNNs) models have been successful in automating this process, the reliance on local textures can negatively impact model performance in the presence of pathological conditions such as brain tumors. This study presents a novel yet practical approach to offer supplementary texture-invariant spatial information of the brain as a geometric prior to enhance shape representations to further enable informed segmentation decisions. Our numerical results demonstrate that the new method outperforms competing algorithms in skull stripping tasks for both healthy and pathological inputs. These findings underscore the potential of incorporating geometric information in deep learning models to enhance the accuracy of brain segmentation.

Original language	English
Title of host publication	Computational Mathematics Modeling in Cancer Analysis
Subtitle of host publication	Second International Workshop, CMMCA 2023, Held in Conjunction with MICCAI 2023, Vancouver, BC, Canada, October 8, 2023, Proceedings
Editors	Wenjian Qin, Nazar Zaki, Fa Zhang, Jia Wu, Fan Yang, Chao Li
Publisher	Springer Nature Switzerland AG
Pages	21-30
Number of pages	10
Edition	1
ISBN (Electronic)	9783031450877
ISBN (Print)	9783031450860
DOIs	https://doi.org/10.1007/978-3-031-45087-7_3
Publication status	Published - 9 Oct 2023
Event	2nd Workshop on Computational Mathematics Modeling in Cancer Analysis, CMMCA 2023 - Vancouver, Canada Duration: 8 Oct 2023 → 8 Oct 2023

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	14243 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	2nd Workshop on Computational Mathematics Modeling in Cancer Analysis, CMMCA 2023
Country/Territory	Canada
City	Vancouver
Period	8/10/23 → 8/10/23

Keywords

Convolutional neural networks
Deep learning
Geometric information
Skull stripping

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-031-45087-7_3

Cite this

Li, Y., Li, C., Wei, Y., Price, S., Schönlieb, C. B., & Chen, X. (2023). G-CNN: Adaptive Geometric Convolutional Neural Networks for MRI-Based Skull Stripping. In W. Qin, N. Zaki, F. Zhang, J. Wu, F. Yang, & C. Li (Eds.), Computational Mathematics Modeling in Cancer Analysis: Second International Workshop, CMMCA 2023, Held in Conjunction with MICCAI 2023, Vancouver, BC, Canada, October 8, 2023, Proceedings (1 ed., pp. 21-30). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 14243 LNCS). Springer Nature Switzerland AG. https://doi.org/10.1007/978-3-031-45087-7_3

Li, Yifan ; Li, Chao ; Wei, Yiran et al. / G-CNN : Adaptive Geometric Convolutional Neural Networks for MRI-Based Skull Stripping. Computational Mathematics Modeling in Cancer Analysis: Second International Workshop, CMMCA 2023, Held in Conjunction with MICCAI 2023, Vancouver, BC, Canada, October 8, 2023, Proceedings. editor / Wenjian Qin ; Nazar Zaki ; Fa Zhang ; Jia Wu ; Fan Yang ; Chao Li. 1. ed. Springer Nature Switzerland AG, 2023. pp. 21-30 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{e68d3f087f8e4a62819271211abeee18,

title = "G-CNN: Adaptive Geometric Convolutional Neural Networks for MRI-Based Skull Stripping",

abstract = "Skull stripping in MRI-based brain imaging involves extraction of brain regions from raw images. While some Convolutional Neural Nets (CNNs) models have been successful in automating this process, the reliance on local textures can negatively impact model performance in the presence of pathological conditions such as brain tumors. This study presents a novel yet practical approach to offer supplementary texture-invariant spatial information of the brain as a geometric prior to enhance shape representations to further enable informed segmentation decisions. Our numerical results demonstrate that the new method outperforms competing algorithms in skull stripping tasks for both healthy and pathological inputs. These findings underscore the potential of incorporating geometric information in deep learning models to enhance the accuracy of brain segmentation.",

keywords = "Convolutional neural networks, Deep learning, Geometric information, Skull stripping",

author = "Yifan Li and Chao Li and Yiran Wei and Stephen Price and Sch{\"o}nlieb, \{Carola Bibiane\} and Xi Chen",

note = "{\textcopyright} The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2023; 2nd Workshop on Computational Mathematics Modeling in Cancer Analysis, CMMCA 2023 ; Conference date: 08-10-2023 Through 08-10-2023",

year = "2023",

month = oct,

day = "9",

doi = "10.1007/978-3-031-45087-7\_3",

language = "English",

isbn = "9783031450860",

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

publisher = "Springer Nature Switzerland AG",

pages = "21--30",

editor = "Wenjian Qin and Nazar Zaki and Fa Zhang and Jia Wu and Fan Yang and Chao Li",

booktitle = "Computational Mathematics Modeling in Cancer Analysis",

address = "Switzerland",

edition = "1",

}

Li, Y, Li, C, Wei, Y, Price, S, Schönlieb, CB & Chen, X 2023, G-CNN: Adaptive Geometric Convolutional Neural Networks for MRI-Based Skull Stripping. in W Qin, N Zaki, F Zhang, J Wu, F Yang & C Li (eds), Computational Mathematics Modeling in Cancer Analysis: Second International Workshop, CMMCA 2023, Held in Conjunction with MICCAI 2023, Vancouver, BC, Canada, October 8, 2023, Proceedings. 1 edn, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 14243 LNCS, Springer Nature Switzerland AG, pp. 21-30, 2nd Workshop on Computational Mathematics Modeling in Cancer Analysis, CMMCA 2023, Vancouver, Canada, 8/10/23. https://doi.org/10.1007/978-3-031-45087-7_3

G-CNN: Adaptive Geometric Convolutional Neural Networks for MRI-Based Skull Stripping. / Li, Yifan; Li, Chao; Wei, Yiran et al.
Computational Mathematics Modeling in Cancer Analysis: Second International Workshop, CMMCA 2023, Held in Conjunction with MICCAI 2023, Vancouver, BC, Canada, October 8, 2023, Proceedings. ed. / Wenjian Qin; Nazar Zaki; Fa Zhang; Jia Wu; Fan Yang; Chao Li. 1. ed. Springer Nature Switzerland AG, 2023. p. 21-30 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 14243 LNCS).

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

TY - GEN

T1 - G-CNN

T2 - 2nd Workshop on Computational Mathematics Modeling in Cancer Analysis, CMMCA 2023

AU - Li, Yifan

AU - Li, Chao

AU - Wei, Yiran

AU - Price, Stephen

AU - Schönlieb, Carola Bibiane

AU - Chen, Xi

N1 - © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2023

PY - 2023/10/9

Y1 - 2023/10/9

N2 - Skull stripping in MRI-based brain imaging involves extraction of brain regions from raw images. While some Convolutional Neural Nets (CNNs) models have been successful in automating this process, the reliance on local textures can negatively impact model performance in the presence of pathological conditions such as brain tumors. This study presents a novel yet practical approach to offer supplementary texture-invariant spatial information of the brain as a geometric prior to enhance shape representations to further enable informed segmentation decisions. Our numerical results demonstrate that the new method outperforms competing algorithms in skull stripping tasks for both healthy and pathological inputs. These findings underscore the potential of incorporating geometric information in deep learning models to enhance the accuracy of brain segmentation.

AB - Skull stripping in MRI-based brain imaging involves extraction of brain regions from raw images. While some Convolutional Neural Nets (CNNs) models have been successful in automating this process, the reliance on local textures can negatively impact model performance in the presence of pathological conditions such as brain tumors. This study presents a novel yet practical approach to offer supplementary texture-invariant spatial information of the brain as a geometric prior to enhance shape representations to further enable informed segmentation decisions. Our numerical results demonstrate that the new method outperforms competing algorithms in skull stripping tasks for both healthy and pathological inputs. These findings underscore the potential of incorporating geometric information in deep learning models to enhance the accuracy of brain segmentation.

KW - Convolutional neural networks

KW - Deep learning

KW - Geometric information

KW - Skull stripping

UR - https://www.scopus.com/pages/publications/85174707928

U2 - 10.1007/978-3-031-45087-7_3

DO - 10.1007/978-3-031-45087-7_3

M3 - Conference contribution

AN - SCOPUS:85174707928

SN - 9783031450860

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

SP - 21

EP - 30

BT - Computational Mathematics Modeling in Cancer Analysis

A2 - Qin, Wenjian

A2 - Zaki, Nazar

A2 - Zhang, Fa

A2 - Wu, Jia

A2 - Yang, Fan

A2 - Li, Chao

PB - Springer Nature Switzerland AG

Y2 - 8 October 2023 through 8 October 2023

ER -

Li Y, Li C, Wei Y, Price S, Schönlieb CB, Chen X. G-CNN: Adaptive Geometric Convolutional Neural Networks for MRI-Based Skull Stripping. In Qin W, Zaki N, Zhang F, Wu J, Yang F, Li C, editors, Computational Mathematics Modeling in Cancer Analysis: Second International Workshop, CMMCA 2023, Held in Conjunction with MICCAI 2023, Vancouver, BC, Canada, October 8, 2023, Proceedings. 1 ed. Springer Nature Switzerland AG. 2023. p. 21-30. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). Epub 2023 Oct 7. doi: 10.1007/978-3-031-45087-7_3