Seg4Reg Networks for Automated Spinal Curvature Estimation

Yi Lin; Hong Yu Zhou; Kai Ma; Xin Yang; Yefeng Zheng

doi:10.1007/978-3-030-39752-4_7

Seg4Reg Networks for Automated Spinal Curvature Estimation

Yi Lin
, Hong Yu Zhou^*
, Kai Ma
, Xin Yang
, Yefeng Zheng

^*Corresponding author for this work

Huazhong University of Science and Technology
Tencent

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

34 Scopus citations

Abstract

In this paper, we propose a new pipeline to perform accurate spinal curvature estimation. The framework, named as Seg4Reg, contains two deep neural networks focusing on segmentation and regression, respectively. Based on the results generated by the segmentation model, the regression network directly predicts the cobb angles from segmentation masks. To alleviate the domain shift problem appeared between training and testing sets, we also conduct a domain adaptation module into network structures. Finally, by ensembling the predictions of different models, our method achieves 21.71 SMAPE in the testing set.

Original language	English
Title of host publication	Computational Methods and Clinical Applications for Spine Imaging - 6th International Workshop and Challenge, CSI 2019, Proceedings
Editors	Yunliang Cai, Liansheng Wang, Michel Audette, Guoyan Zheng, Shuo Li
Publisher	Springer
Pages	69-74
Number of pages	6
ISBN (Print)	9783030397517
DOIs	https://doi.org/10.1007/978-3-030-39752-4_7
State	Published - 2020
Externally published	Yes
Event	6th International Workshop and Challenge on Computational Methods and Clinical Applications for Spine Imaging, CSI 2019, held in conjunction with the 22nd International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2019 - Shenzhen, China Duration: 17 Oct 2019 → 17 Oct 2019

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	11963 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	6th International Workshop and Challenge on Computational Methods and Clinical Applications for Spine Imaging, CSI 2019, held in conjunction with the 22nd International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2019
Country/Territory	China
City	Shenzhen
Period	17/10/19 → 17/10/19

Keywords

Cobb angle
Deep learning
Spinal curvature estimation

Access to Document

10.1007/978-3-030-39752-4_7

Cite this

Lin, Y., Zhou, H. Y., Ma, K., Yang, X., & Zheng, Y. (2020). Seg4Reg Networks for Automated Spinal Curvature Estimation. In Y. Cai, L. Wang, M. Audette, G. Zheng, & S. Li (Eds.), Computational Methods and Clinical Applications for Spine Imaging - 6th International Workshop and Challenge, CSI 2019, Proceedings (pp. 69-74). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11963 LNCS). Springer. https://doi.org/10.1007/978-3-030-39752-4_7

Lin, Yi ; Zhou, Hong Yu ; Ma, Kai et al. / Seg4Reg Networks for Automated Spinal Curvature Estimation. Computational Methods and Clinical Applications for Spine Imaging - 6th International Workshop and Challenge, CSI 2019, Proceedings. editor / Yunliang Cai ; Liansheng Wang ; Michel Audette ; Guoyan Zheng ; Shuo Li. Springer, 2020. pp. 69-74 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{d7b53219eacb46a88ae5f9970b5d82aa,

title = "Seg4Reg Networks for Automated Spinal Curvature Estimation",

abstract = "In this paper, we propose a new pipeline to perform accurate spinal curvature estimation. The framework, named as Seg4Reg, contains two deep neural networks focusing on segmentation and regression, respectively. Based on the results generated by the segmentation model, the regression network directly predicts the cobb angles from segmentation masks. To alleviate the domain shift problem appeared between training and testing sets, we also conduct a domain adaptation module into network structures. Finally, by ensembling the predictions of different models, our method achieves 21.71 SMAPE in the testing set.",

keywords = "Cobb angle, Deep learning, Spinal curvature estimation",

author = "Yi Lin and Zhou, \{Hong Yu\} and Kai Ma and Xin Yang and Yefeng Zheng",

note = "Publisher Copyright: {\textcopyright} Springer Nature Switzerland AG 2020.; 6th International Workshop and Challenge on Computational Methods and Clinical Applications for Spine Imaging, CSI 2019, held in conjunction with the 22nd International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2019 ; Conference date: 17-10-2019 Through 17-10-2019",

year = "2020",

doi = "10.1007/978-3-030-39752-4\_7",

language = "英语",

isbn = "9783030397517",

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

publisher = "Springer",

pages = "69--74",

editor = "Yunliang Cai and Liansheng Wang and Michel Audette and Guoyan Zheng and Shuo Li",

booktitle = "Computational Methods and Clinical Applications for Spine Imaging - 6th International Workshop and Challenge, CSI 2019, Proceedings",

address = "德国",

}

Lin, Y, Zhou, HY, Ma, K, Yang, X & Zheng, Y 2020, Seg4Reg Networks for Automated Spinal Curvature Estimation. in Y Cai, L Wang, M Audette, G Zheng & S Li (eds), Computational Methods and Clinical Applications for Spine Imaging - 6th International Workshop and Challenge, CSI 2019, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11963 LNCS, Springer, pp. 69-74, 6th International Workshop and Challenge on Computational Methods and Clinical Applications for Spine Imaging, CSI 2019, held in conjunction with the 22nd International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2019, Shenzhen, China, 17/10/19. https://doi.org/10.1007/978-3-030-39752-4_7

Seg4Reg Networks for Automated Spinal Curvature Estimation. / Lin, Yi; Zhou, Hong Yu; Ma, Kai et al.
Computational Methods and Clinical Applications for Spine Imaging - 6th International Workshop and Challenge, CSI 2019, Proceedings. ed. / Yunliang Cai; Liansheng Wang; Michel Audette; Guoyan Zheng; Shuo Li. Springer, 2020. p. 69-74 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11963 LNCS).

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

TY - GEN

T1 - Seg4Reg Networks for Automated Spinal Curvature Estimation

AU - Lin, Yi

AU - Zhou, Hong Yu

AU - Ma, Kai

AU - Yang, Xin

AU - Zheng, Yefeng

N1 - Publisher Copyright: © Springer Nature Switzerland AG 2020.

PY - 2020

Y1 - 2020

N2 - In this paper, we propose a new pipeline to perform accurate spinal curvature estimation. The framework, named as Seg4Reg, contains two deep neural networks focusing on segmentation and regression, respectively. Based on the results generated by the segmentation model, the regression network directly predicts the cobb angles from segmentation masks. To alleviate the domain shift problem appeared between training and testing sets, we also conduct a domain adaptation module into network structures. Finally, by ensembling the predictions of different models, our method achieves 21.71 SMAPE in the testing set.

AB - In this paper, we propose a new pipeline to perform accurate spinal curvature estimation. The framework, named as Seg4Reg, contains two deep neural networks focusing on segmentation and regression, respectively. Based on the results generated by the segmentation model, the regression network directly predicts the cobb angles from segmentation masks. To alleviate the domain shift problem appeared between training and testing sets, we also conduct a domain adaptation module into network structures. Finally, by ensembling the predictions of different models, our method achieves 21.71 SMAPE in the testing set.

KW - Cobb angle

KW - Deep learning

KW - Spinal curvature estimation

UR - https://www.scopus.com/pages/publications/85080919518

U2 - 10.1007/978-3-030-39752-4_7

DO - 10.1007/978-3-030-39752-4_7

M3 - 会议稿件

AN - SCOPUS:85080919518

SN - 9783030397517

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

SP - 69

EP - 74

BT - Computational Methods and Clinical Applications for Spine Imaging - 6th International Workshop and Challenge, CSI 2019, Proceedings

A2 - Cai, Yunliang

A2 - Wang, Liansheng

A2 - Audette, Michel

A2 - Zheng, Guoyan

A2 - Li, Shuo

PB - Springer

T2 - 6th International Workshop and Challenge on Computational Methods and Clinical Applications for Spine Imaging, CSI 2019, held in conjunction with the 22nd International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2019

Y2 - 17 October 2019 through 17 October 2019

ER -

Lin Y, Zhou HY, Ma K, Yang X, Zheng Y. Seg4Reg Networks for Automated Spinal Curvature Estimation. In Cai Y, Wang L, Audette M, Zheng G, Li S, editors, Computational Methods and Clinical Applications for Spine Imaging - 6th International Workshop and Challenge, CSI 2019, Proceedings. Springer. 2020. p. 69-74. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-39752-4_7

Seg4Reg Networks for Automated Spinal Curvature Estimation

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this