Statistical Dependency Guided Contrastive Learning for Multiple Labeling in Prenatal Ultrasound

Shuangchi He; Zehui Lin; Xin Yang; Chaoyu Chen; Jian Wang; Xue Shuang; Ziwei Deng; Qin Liu; Yan Cao; Xiduo Lu; Ruobing Huang; Nishant Ravikumar; Alejandro F Frangi; Yuanji Zhang; Yi Xiong; Dong Ni

doi:10.1007/978-3-030-87589-3_20

Statistical Dependency Guided Contrastive Learning for Multiple Labeling in Prenatal Ultrasound

Shuangchi He, Zehui Lin, Xin Yang, Chaoyu Chen, Jian Wang, Xue Shuang, Ziwei Deng, Qin Liu, Yan Cao, Xiduo Lu, Ruobing Huang, Nishant Ravikumar, Alejandro F Frangi, Yuanji Zhang, Yi Xiong, Dong Ni^*

^*Corresponding author for this work

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

8 Downloads (Pure)

Abstract

Standard plane recognition plays an important role in prenatal ultrasound (US) screening. Automatically recognizing the standard plane along with the corresponding anatomical structures in US image can not only facilitate US image interpretation but also improve diagnostic efficiency. In this study, we build a novel multi-label learning (MLL) scheme to identify multiple standard planes and corresponding anatomical structures of fetus simultaneously. Our contribution is three-fold. First, we represent the class correlation by word embeddings to capture the fine-grained semantic and latent statistical concurrency. Second, we equip the MLL with a graph convolutional network to explore the inner and outer relationship among categories. Third, we propose a novel cluster relabel-based contrastive learning algorithm to encourage the divergence among ambiguous classes. Extensive validation was performed on our large in-house dataset. Our approach reports the highest accuracy as 90.25 % for standard planes labeling, 85.59 % for planes and structures labeling and mAP as 94.63 %. The proposed MLL scheme provides a novel perspective for standard plane recognition and can be easily extended to other medical image classification tasks.

Original language	English
Title of host publication	Machine Learning in Medical Imaging - 12th International Workshop, MLMI 2021, Held in Conjunction with MICCAI 2021, Proceedings
Editors	Chunfeng Lian, Xiaohuan Cao, Islem Rekik, Xuanang Xu, Pingkun Yan
Publisher	Springer Nature
Pages	190-198
Number of pages	9
ISBN (Print)	9783030875886
DOIs	https://doi.org/10.1007/978-3-030-87589-3_20
Publication status	Published - 21 Sept 2021
Event	12th International Workshop on Machine Learning in Medical Imaging, MLMI 2021, held in conjunction with 24th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2021 - Virtual, Online Duration: 27 Sept 2021 → 27 Sept 2021

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	12966 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	12th International Workshop on Machine Learning in Medical Imaging, MLMI 2021, held in conjunction with 24th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2021
City	Virtual, Online
Period	27/09/21 → 27/09/21

Keywords

cs.CV

Access to Document

10.1007/978-3-030-87589-3_20

2108.05055v3

Cite this

He, S., Lin, Z., Yang, X., Chen, C., Wang, J., Shuang, X., Deng, Z., Liu, Q., Cao, Y., Lu, X., Huang, R., Ravikumar, N., Frangi, A. F., Zhang, Y., Xiong, Y., & Ni, D. (2021). Statistical Dependency Guided Contrastive Learning for Multiple Labeling in Prenatal Ultrasound. In C. Lian, X. Cao, I. Rekik, X. Xu, & P. Yan (Eds.), Machine Learning in Medical Imaging - 12th International Workshop, MLMI 2021, Held in Conjunction with MICCAI 2021, Proceedings (pp. 190-198). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12966 LNCS). Springer Nature. https://doi.org/10.1007/978-3-030-87589-3_20

He, Shuangchi ; Lin, Zehui ; Yang, Xin et al. / Statistical Dependency Guided Contrastive Learning for Multiple Labeling in Prenatal Ultrasound. Machine Learning in Medical Imaging - 12th International Workshop, MLMI 2021, Held in Conjunction with MICCAI 2021, Proceedings. editor / Chunfeng Lian ; Xiaohuan Cao ; Islem Rekik ; Xuanang Xu ; Pingkun Yan. Springer Nature, 2021. pp. 190-198 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{50970494d31c463fa9c64b2381f54fd9,

title = "Statistical Dependency Guided Contrastive Learning for Multiple Labeling in Prenatal Ultrasound",

abstract = "Standard plane recognition plays an important role in prenatal ultrasound (US) screening. Automatically recognizing the standard plane along with the corresponding anatomical structures in US image can not only facilitate US image interpretation but also improve diagnostic efficiency. In this study, we build a novel multi-label learning (MLL) scheme to identify multiple standard planes and corresponding anatomical structures of fetus simultaneously. Our contribution is three-fold. First, we represent the class correlation by word embeddings to capture the fine-grained semantic and latent statistical concurrency. Second, we equip the MLL with a graph convolutional network to explore the inner and outer relationship among categories. Third, we propose a novel cluster relabel-based contrastive learning algorithm to encourage the divergence among ambiguous classes. Extensive validation was performed on our large in-house dataset. Our approach reports the highest accuracy as 90.25 % for standard planes labeling, 85.59 % for planes and structures labeling and mAP as 94.63 %. The proposed MLL scheme provides a novel perspective for standard plane recognition and can be easily extended to other medical image classification tasks.",

keywords = "cs.CV",

author = "Shuangchi He and Zehui Lin and Xin Yang and Chaoyu Chen and Jian Wang and Xue Shuang and Ziwei Deng and Qin Liu and Yan Cao and Xiduo Lu and Ruobing Huang and Nishant Ravikumar and Frangi, {Alejandro F} and Yuanji Zhang and Yi Xiong and Dong Ni",

note = "Funding Information: This work was supported by the SZU Top Ranking Project (No. Funding Information: This work was supported by the SZU Top Ranking Project (No. 86000000210). Publisher Copyright: {\textcopyright} 2021, Springer Nature Switzerland AG.; 12th International Workshop on Machine Learning in Medical Imaging, MLMI 2021, held in conjunction with 24th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2021 ; Conference date: 27-09-2021 Through 27-09-2021",

year = "2021",

month = sep,

day = "21",

doi = "10.1007/978-3-030-87589-3_20",

language = "English",

isbn = "9783030875886",

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

publisher = "Springer Nature",

pages = "190--198",

editor = "Chunfeng Lian and Xiaohuan Cao and Islem Rekik and Xuanang Xu and Pingkun Yan",

booktitle = "Machine Learning in Medical Imaging - 12th International Workshop, MLMI 2021, Held in Conjunction with MICCAI 2021, Proceedings",

address = "United States",

}

He, S, Lin, Z, Yang, X, Chen, C, Wang, J, Shuang, X, Deng, Z, Liu, Q, Cao, Y, Lu, X, Huang, R, Ravikumar, N, Frangi, AF, Zhang, Y, Xiong, Y & Ni, D 2021, Statistical Dependency Guided Contrastive Learning for Multiple Labeling in Prenatal Ultrasound. in C Lian, X Cao, I Rekik, X Xu & P Yan (eds), Machine Learning in Medical Imaging - 12th International Workshop, MLMI 2021, Held in Conjunction with MICCAI 2021, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 12966 LNCS, Springer Nature, pp. 190-198, 12th International Workshop on Machine Learning in Medical Imaging, MLMI 2021, held in conjunction with 24th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2021, Virtual, Online, 27/09/21. https://doi.org/10.1007/978-3-030-87589-3_20

Statistical Dependency Guided Contrastive Learning for Multiple Labeling in Prenatal Ultrasound. / He, Shuangchi; Lin, Zehui; Yang, Xin et al.
Machine Learning in Medical Imaging - 12th International Workshop, MLMI 2021, Held in Conjunction with MICCAI 2021, Proceedings. ed. / Chunfeng Lian; Xiaohuan Cao; Islem Rekik; Xuanang Xu; Pingkun Yan. Springer Nature, 2021. p. 190-198 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12966 LNCS).

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

TY - GEN

T1 - Statistical Dependency Guided Contrastive Learning for Multiple Labeling in Prenatal Ultrasound

AU - He, Shuangchi

AU - Lin, Zehui

AU - Yang, Xin

AU - Chen, Chaoyu

AU - Wang, Jian

AU - Shuang, Xue

AU - Deng, Ziwei

AU - Liu, Qin

AU - Cao, Yan

AU - Lu, Xiduo

AU - Huang, Ruobing

AU - Ravikumar, Nishant

AU - Frangi, Alejandro F

AU - Zhang, Yuanji

AU - Xiong, Yi

AU - Ni, Dong

N1 - Funding Information: This work was supported by the SZU Top Ranking Project (No. Funding Information: This work was supported by the SZU Top Ranking Project (No. 86000000210). Publisher Copyright: © 2021, Springer Nature Switzerland AG.

PY - 2021/9/21

Y1 - 2021/9/21

N2 - Standard plane recognition plays an important role in prenatal ultrasound (US) screening. Automatically recognizing the standard plane along with the corresponding anatomical structures in US image can not only facilitate US image interpretation but also improve diagnostic efficiency. In this study, we build a novel multi-label learning (MLL) scheme to identify multiple standard planes and corresponding anatomical structures of fetus simultaneously. Our contribution is three-fold. First, we represent the class correlation by word embeddings to capture the fine-grained semantic and latent statistical concurrency. Second, we equip the MLL with a graph convolutional network to explore the inner and outer relationship among categories. Third, we propose a novel cluster relabel-based contrastive learning algorithm to encourage the divergence among ambiguous classes. Extensive validation was performed on our large in-house dataset. Our approach reports the highest accuracy as 90.25 % for standard planes labeling, 85.59 % for planes and structures labeling and mAP as 94.63 %. The proposed MLL scheme provides a novel perspective for standard plane recognition and can be easily extended to other medical image classification tasks.

AB - Standard plane recognition plays an important role in prenatal ultrasound (US) screening. Automatically recognizing the standard plane along with the corresponding anatomical structures in US image can not only facilitate US image interpretation but also improve diagnostic efficiency. In this study, we build a novel multi-label learning (MLL) scheme to identify multiple standard planes and corresponding anatomical structures of fetus simultaneously. Our contribution is three-fold. First, we represent the class correlation by word embeddings to capture the fine-grained semantic and latent statistical concurrency. Second, we equip the MLL with a graph convolutional network to explore the inner and outer relationship among categories. Third, we propose a novel cluster relabel-based contrastive learning algorithm to encourage the divergence among ambiguous classes. Extensive validation was performed on our large in-house dataset. Our approach reports the highest accuracy as 90.25 % for standard planes labeling, 85.59 % for planes and structures labeling and mAP as 94.63 %. The proposed MLL scheme provides a novel perspective for standard plane recognition and can be easily extended to other medical image classification tasks.

KW - cs.CV

UR - http://www.scopus.com/inward/record.url?scp=85116480518&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-87589-3_20

DO - 10.1007/978-3-030-87589-3_20

M3 - Conference contribution

SN - 9783030875886

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

SP - 190

EP - 198

BT - Machine Learning in Medical Imaging - 12th International Workshop, MLMI 2021, Held in Conjunction with MICCAI 2021, Proceedings

A2 - Lian, Chunfeng

A2 - Cao, Xiaohuan

A2 - Rekik, Islem

A2 - Xu, Xuanang

A2 - Yan, Pingkun

PB - Springer Nature

T2 - 12th International Workshop on Machine Learning in Medical Imaging, MLMI 2021, held in conjunction with 24th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2021

Y2 - 27 September 2021 through 27 September 2021

ER -

He S, Lin Z, Yang X, Chen C, Wang J, Shuang X et al. Statistical Dependency Guided Contrastive Learning for Multiple Labeling in Prenatal Ultrasound. In Lian C, Cao X, Rekik I, Xu X, Yan P, editors, Machine Learning in Medical Imaging - 12th International Workshop, MLMI 2021, Held in Conjunction with MICCAI 2021, Proceedings. Springer Nature. 2021. p. 190-198. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). Epub 2021 Sept 21. doi: 10.1007/978-3-030-87589-3_20

Statistical Dependency Guided Contrastive Learning for Multiple Labeling in Prenatal Ultrasound

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this