Towards Better Surgical Instrument Segmentation in Endoscopic Vision: Multi-Angle Feature Aggregation and Contour Supervision

doi:10.1109/LRA.2020.3009073

CASIA OpenIR > 中科院工业视觉智能装备工程实验室 > 精密感知与控制

	Towards Better Surgical Instrument Segmentation in Endoscopic Vision: Multi-Angle Feature Aggregation and Contour Supervision
	Qin, Fangbo1 ; Lin, Shan 2; Li, Yangming 4; Bly, Randall A.3; Moe, Kris S.3; Hannaford, Blake 2
发表期刊	IEEE ROBOTICS AND AUTOMATION LETTERS
ISSN	2377-3766
	2020-10-01
卷号	5 期号:4 页码:6639-6646
通讯作者	Hannaford, Blake(blake@uw.edu)
摘要	Accurate and real-time surgical instrument segmentation is important in the endoscopic vision of robot-assisted surgery, and significant challenges are posed by frequent instrument-tissue contacts and continuous change of observation perspective. For these challenging tasks more and more deep neural networks (DNN) models are designed in recent years. We are motivated to propose a general embeddable approach to improve these current DNN segmentation models without increasing the model parameter number. Firstly, observing the limited rotation-invariance performance of DNN, we proposed the Multi-Angle Feature Aggregation (MAFA) method, leveraging active image rotation to gain richer visual cues and make the prediction more robust to instrument orientation changes. Secondly, in the end-to-end training stage, the auxiliary contour supervision is utilized to guide the model to learn the boundary awareness, so that the contour shape of segmentation mask is more precise. The proposed method is validated with ablation experiments on the novel Sinus-Surgery datasets collected from surgeons' operations, and is compared to the existing methods on a public dataset collected with a da Vinci Xi Robot.
关键词	Computer vision for medical robotics medical robots and systems deep learning for visual perception object detection segmentation and categorization
DOI	10.1109/LRA.2020.3009073
关键词[WOS]	SURGERY ; ROBOTICS
收录类别	SCI
语种	英语
资助项目	National Science Foundation[IIS-1637444] ; National Natural Science Foundation of China[61703398]
项目资助者	National Science Foundation ; National Natural Science Foundation of China
WOS研究方向	Robotics
WOS类目	Robotics
WOS记录号	WOS:000564288600005
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
七大方向——子方向分类	人工智能+医疗
引用统计	被引频次：36[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/41530
专题	中科院工业视觉智能装备工程实验室_精密感知与控制
通讯作者	Hannaford, Blake
作者单位	1.Chinese Acad Sci, Inst Automat, Res Ctr Precis Sensing & Control, Beijing 100190, Peoples R China 2.Univ Washington UW, Dept Elect Engn, Seattle, WA 98195 USA 3.Univ Washington, Dept Otolaryngol Head & Neck Surg, Seattle, WA 98195 USA 4.Rochester Inst Technol, Dept Elect Comp & Telecommun Engn Technol, Rochester, NY 14623 USA
第一作者单位	精密感知与控制研究中心
推荐引用方式 GB/T 7714	Qin, Fangbo,Lin, Shan,Li, Yangming,et al. Towards Better Surgical Instrument Segmentation in Endoscopic Vision: Multi-Angle Feature Aggregation and Contour Supervision[J]. IEEE ROBOTICS AND AUTOMATION LETTERS,2020,5(4):6639-6646.
APA	Qin, Fangbo,Lin, Shan,Li, Yangming,Bly, Randall A.,Moe, Kris S.,&Hannaford, Blake.(2020).Towards Better Surgical Instrument Segmentation in Endoscopic Vision: Multi-Angle Feature Aggregation and Contour Supervision.IEEE ROBOTICS AND AUTOMATION LETTERS,5(4),6639-6646.
MLA	Qin, Fangbo,et al."Towards Better Surgical Instrument Segmentation in Endoscopic Vision: Multi-Angle Feature Aggregation and Contour Supervision".IEEE ROBOTICS AND AUTOMATION LETTERS 5.4(2020):6639-6646.