中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 自动化研究所
  3. 中国科学院自动化研究所
  4. 复杂系统认知与决策实验室
  5. 先进机器人控制团队
SurgiNet: Pyramid Attention Aggregation and Class-wise Self-Distillation for Surgical Instrument Segmentation

文献类型：期刊论文

作者Ni, Zhen-Liang1,2; Zhou, Xiao-Hu1; Wang, Guan-An1,2; Yue, Wen-Qian1; Li, Zhen1; Bian, Gui-Bin1,2; Hou, Zeng-Guang1,2,3
刊名MEDICAL IMAGE ANALYSIS
出版日期2022-02-01
卷号76页码:10
ISSN号1361-8415
关键词Surgical Insturment Segmentation Class-wise Self-Distillation Pyramid Attention
DOI10.1016/j.media.2021.102310
通讯作者Bian, Gui-Bin(guibin.bian@ia.ac.cn) ; Hou, Zeng-Guang(zengguang.hou@ia.ac.cn)
英文摘要Surgical instrument segmentation plays a promising role in robot-assisted surgery. However, illumination issues often appear in surgical scenes, altering the color and texture of surgical instruments. Changes in visual features make surgical instrument segmentation difficult. To address illumination issues, the SurgiNet is proposed to learn pyramid attention features. The double attention module is designed to capture the semantic dependencies between locations and channels. Based on semantic dependencies, the semantic features in the disturbed area can be inferred for addressing illumination issues. Pyramid attention is aggregated to capture multi-scale features and make predictions more accurate. To perform model compression, class-wise self-distillation is proposed to enhance the representation learning of the network, which performs feature distillation within the class to eliminate interference from other classes. Top-down and multi-stage knowledge distillation is designed to distill class probability maps. By inter layer supervision, high-level probability maps are applied to calibrate the probability distribution of lowlevel probability maps. Since class-wise distillation enhances the self-learning of the network, the network can get excellent performance with a lightweight backbone. The proposed network achieves the state-of-the-art performance of 89.14% mIoU on CataIS with only 1.66 GFlops and 2.05 M parameters. It also takes first place on EndoVis 2017 with 66.30% mIoU. (c) 2021 Published by Elsevier B.V.
资助项目National Natural Science Foundation of China[62027813] ; National Natural Science Foundation of China[62003343] ; National Natural Science Foundation of China[U1913210] ; National Natural Science Foundation of China[U1913601] ; National Natural Science Foundation of China[U1713220] ; Beijing Science and Technology Star[Z19110 0 0 01119046] ; Youth Innovation Promotion Association of the Chinese Academy of Sciences[2018165] ; Youth Innovation Promotion Association of the Chinese Academy of Sciences[2020140]
WOS研究方向Computer Science ; Engineering ; Radiology, Nuclear Medicine & Medical Imaging
语种英语
出版者ELSEVIER
WOS记录号WOS:000742838900002
资助机构National Natural Science Foundation of China ; Beijing Science and Technology Star ; Youth Innovation Promotion Association of the Chinese Academy of Sciences
源URL[http://ir.ia.ac.cn/handle/173211/47042]  
专题自动化研究所_复杂系统管理与控制国家重点实验室_先进机器人控制团队
通讯作者Bian, Gui-Bin; Hou, Zeng-Guang
作者单位1.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing, Peoples R China
2.Chinese Acad Sci, Inst Automat, State Key Lab Management & Control Complex Syst, Beijing 100190, Peoples R China
3.Macau Univ Sci & Technol, Inst Syst Engn, Joint Lab Intelligence Sci & Technol, Macau, Peoples R China
推荐引用方式
GB/T 7714		 Ni, Zhen-Liang,Zhou, Xiao-Hu,Wang, Guan-An,et al. SurgiNet: Pyramid Attention Aggregation and Class-wise Self-Distillation for Surgical Instrument Segmentation[J]. MEDICAL IMAGE ANALYSIS,2022,76:10.
APA Ni, Zhen-Liang.,Zhou, Xiao-Hu.,Wang, Guan-An.,Yue, Wen-Qian.,Li, Zhen.,...&Hou, Zeng-Guang.(2022).SurgiNet: Pyramid Attention Aggregation and Class-wise Self-Distillation for Surgical Instrument Segmentation.MEDICAL IMAGE ANALYSIS,76,10.
MLA Ni, Zhen-Liang,et al."SurgiNet: Pyramid Attention Aggregation and Class-wise Self-Distillation for Surgical Instrument Segmentation".MEDICAL IMAGE ANALYSIS 76(2022):10.

入库方式： OAI收割

来源：自动化研究所

浏览0
下载0
收藏0
其他版本

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。