BiMNet: A Multimodal Data Fusion Network for continuous circular capsulorhexis Action Segmentation

doi:10.1016/j.eswa.2023.121885

	BiMNet: A Multimodal Data Fusion Network for continuous circular capsulorhexis Action Segmentation
	Bian, Gui-Bin 4,5; Zheng, Jia-Ying 4,5; Li, Zhen 4; Wang, Jie 4,5; Fu, Pan 4,5; Xin, Chen 3; da Silva, Daniel Santos 2; Wu, Wan-Qing 1; De Albuquerque, Victor Hugo C.2
刊名	EXPERT SYSTEMS WITH APPLICATIONS
	2024-03-15
卷号	238 页码:10
关键词	Cataract surgery Continuous circumferential capsulotomy Continuous action segmentation Multimodal data fusion Imbalanced data
ISSN号	0957-4174
DOI	10.1016/j.eswa.2023.121885
通讯作者	Bian, Gui-Bin(guibin.bian@ia.ac.cn)
英文摘要	Completing continuous circular capsulorhexis (CCC) requires the operator to perform fine operations, which is difficult to do accurately when continuous fine actions are out of balance in the classification of CCC procedures. Multimodal deep learning can improve the classifier's performance, but the recognition accuracy of inferior classes is difficult to improve. To solve these problems, a bidirect-gate recurrent unit (Bi-GRU)-attention-based multimodal, multi-timescale data fusion network (BiMNet) is proposed, which contains a data extraction module called a skip-concatenate gate recurrent unit (SC-GRU), a bimodal data fusion attention computation, and a decoder module. The combination of these modules can fully extract the features of different temporal scales in multimodal action data and fuse them effectively. The model is validated using the ophthalmologist CCC multimodal maneuver dataset, which was collected by the data collection platform constructed in this research, achieving an accuracy of 0.9124 +/- 0.0125 in continuous action sequence segmentation and improving the F1-score of minority class recognition to over 80%, making it more effective than baseline algorithms.
资助项目	National Natural Science Foun-dation of China[62027813] ; National Natural Science Foun-dation of China[U20A20196] ; National Key Re-search and Development Program of China[2022YFB4702900] ; Beijing Science Fund for Distinguished Young Scholars, China[JQ21016] ; Excellent member of CAS Youth Innovation Promotion Association, China[Y2022054]
WOS研究方向	Computer Science ; Engineering ; Operations Research & Management Science
语种	英语
出版者	PERGAMON-ELSEVIER SCIENCE LTD
WOS记录号	WOS:001088900900001
资助机构	National Natural Science Foun-dation of China ; National Key Re-search and Development Program of China ; Beijing Science Fund for Distinguished Young Scholars, China ; Excellent member of CAS Youth Innovation Promotion Association, China
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/54292]
专题	智能机器人系统研究
通讯作者	Bian, Gui-Bin
作者单位	1.Sun Yat Sen Univ, Sch Biomed Engn, Guangzhou 510275, Peoples R China 2.Univ Fed Ceara, Dept Teleinformat Engn, BR-60811905 Fortaleza, CE, Brazil 3.Capital Med Univ, Beijing Tongren Hosp, Ophthalmol Dept, Beijing 100005, Peoples R China 4.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China 5.Beijing Informat Sci & Technol Univ, Sch Automat, Beijing 100096, Peoples R China
推荐引用方式 GB/T 7714	Bian, Gui-Bin,Zheng, Jia-Ying,Li, Zhen,et al. BiMNet: A Multimodal Data Fusion Network for continuous circular capsulorhexis Action Segmentation[J]. EXPERT SYSTEMS WITH APPLICATIONS,2024,238:10.
APA	Bian, Gui-Bin.,Zheng, Jia-Ying.,Li, Zhen.,Wang, Jie.,Fu, Pan.,...&De Albuquerque, Victor Hugo C..(2024).BiMNet: A Multimodal Data Fusion Network for continuous circular capsulorhexis Action Segmentation.EXPERT SYSTEMS WITH APPLICATIONS,238,10.
MLA	Bian, Gui-Bin,et al."BiMNet: A Multimodal Data Fusion Network for continuous circular capsulorhexis Action Segmentation".EXPERT SYSTEMS WITH APPLICATIONS 238(2024):10.