Comparative validation of multi-instance instrument segmentation in endoscopy: Results of the ROBUST-MIS 2019 challenge

doi:10.1016/j.media.2020.101920

CORC > 自动化研究所 > 中国科学院自动化研究所 > 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

	Comparative validation of multi-instance instrument segmentation in endoscopy: Results of the ROBUST-MIS 2019 challenge
	Ross, Tobias 3,4; Reinke, Annika 3,4; Full, Peter M.1,4; Wagner, Martin 5; Kenngott, Hannes 5; Apitz, Martin 5; Hempe, Hellena 3; Mindroc-Filimon, Diana 3; Scholz, Patrick 3,6; Thuy Nuong Tran 3
刊名	MEDICAL IMAGE ANALYSIS
	2021-05-01
卷号	70 页码:26
关键词	Multi-instance instrument Minimally invasive surgery Robustness and generalization Surgical data science
ISSN号	1361-8415
DOI	10.1016/j.media.2020.101920
通讯作者	Ross, Tobias(t.ross@dkfz-heidelberg.de)
英文摘要	when run on challenging images (e.g. in the presence of blood, smoke or motion artifacts). Secondly, generalization ; algorithms trained for a specific intervention in a specific hospital should generalize to other interventions or institutions. In an effort to promote solutions for these limitations, we organized the Robust Medical Instrument Segmentation (ROBUST-MIS) challenge as an international benchmarking competition with a specific focus on the robustness and generalization capabilities of algorithms. For the first time in the field of endoscopic image processing, our challenge included a task on binary segmentation and also addressed multiinstance detection and segmentation. The challenge was based on a surgical data set comprising 10,040 annotated images acquired from a total of 30 surgical procedures from three different types of surgery. The validation of the competing methods for the three tasks (binary segmentation, multi-instance detection and multi-instance segmentation) was performed in three different stages with an increasing domain gap between the training and the test data. The results confirm the initial hypothesis, namely that algo Intraoperative tracking of laparoscopic instruments is often a prerequisite for computer and robotic assisted interventions. While numerous methods for detecting, segmenting and tracking of medical instruments based on endoscopic video images have been proposed in the literature, key limitations remain to be addressed: Firstly, robustness , that is, the reliable performance of state-of-the-art methods when run on challenging images (e.g. in the presence of blood, smoke or motion artifacts). Secondly, generalization ; algorithms trained for a specific intervention in a specific hospital should generalize to other interventions or institutions. In an effort to promote solutions for these limitations, we organized the Robust Medical Instrument Segmentation (ROBUST-MIS) challenge as an international benchmarking competition with a specific focus on the robustness and generalization capabilities of algorithms. For the first time in the field of endoscopic image processing, our challenge included a task on binary segmentation and also addressed multi instance detection and segmentation. The challenge was based on a surgical data set comprising 10,040 annotated images acquired from a total of 30 surgical procedures from three different types of surgery. The validation of the competing methods for the three tasks (binary segmentation, multi-instance detection and multi-instance segmentation) was performed in three different stages with an increasing domain gap between the training and the test data. The results confirm the initial hypothesis, namely that algorithm performance degrades with an increasing domain gap. While the average detection and segmentation quality of the best-performing algorithms is high, future research should concentrate on detection and segmentation of small, crossing, moving and transparent instrument(s) (parts). (c) 2020 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license ( http://creativecommons.org/licenses/by-nc-nd/4.0/ )
资助项目	Surgical Oncology Program of the National Center for Tumor Diseases (NCT) Heidelberg ; German Federal Ministry of Economic Affairs and Energy[BMWI 01MT17001C] ; German Federal Ministry of Economic Affairs and Energy[BMWI 01MD15002E] ; UNDERSTAND.AI ; NVIDIA GmbH ; Digital Surgery ; Helmholtz Association under the joint research school HIDSS4Health (Helmholtz Information and Data Science School for Health) ; National Key Research and Development Program of China[2017YFB1302704] ; National Natural Science Fundation of China[81771921] ; National Natural Science Fundation of China[61901084] ; Research Council of Norway[263,248] ; FWF Austrian Science Fund[P 32010-N38]
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:000639613600009
资助机构	Surgical Oncology Program of the National Center for Tumor Diseases (NCT) Heidelberg ; German Federal Ministry of Economic Affairs and Energy ; UNDERSTAND.AI ; NVIDIA GmbH ; Digital Surgery ; Helmholtz Association under the joint research school HIDSS4Health (Helmholtz Information and Data Science School for Health) ; National Key Research and Development Program of China ; National Natural Science Fundation of China ; Research Council of Norway ; FWF Austrian Science Fund
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/44539]
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	Ross, Tobias
作者单位	1.Div Med Image Comp MIC, Neuenheimer Feld 223, D-69120 Heidelberg, Germany 2.Univ Chinese Acad Sci, 52 Sanlihe Rd, Beijing, Peoples R China 3.German Canc Res Ctr, Comp Assisted Med Intervent CAMI, Neuenheimer Feld 223, D-69120 Heidelberg, Germany 4.Heidelberg Univ, Seminarstr 2, D-69117 Heidelberg, Germany 5.Heidelberg Univ Hosp, Dept Gen Visceral & Transplantat Surg, Neuenheimer Feld 110, D-69120 Heidelberg, Germany 6.HIDSS4Hlth Helmholtz Informat & Data Sci Sch Hlth, Neuenheimer Feld 223, D-69120 Heidelberg, Germany 7.Univ Los Andes, Cra 1 18A-12, Bogota 111711, Colombia 8.Chinese Acad Sci, Inst Automat, State Key Lab Management & Control Complex Syst, Beijing 100864, Peoples R China 9.Natl Ctr Tumor Dis NCT, Partner Site Dresden, Dresden, Germany 10.German Canc Res Ctr, Neuenheimer Feld 460, D-69120 Heidelberg, Germany
推荐引用方式 GB/T 7714	Ross, Tobias,Reinke, Annika,Full, Peter M.,et al. Comparative validation of multi-instance instrument segmentation in endoscopy: Results of the ROBUST-MIS 2019 challenge[J]. MEDICAL IMAGE ANALYSIS,2021,70:26.
APA	Ross, Tobias.,Reinke, Annika.,Full, Peter M..,Wagner, Martin.,Kenngott, Hannes.,...&Maier-Hein, Lena.(2021).Comparative validation of multi-instance instrument segmentation in endoscopy: Results of the ROBUST-MIS 2019 challenge.MEDICAL IMAGE ANALYSIS,70,26.
MLA	Ross, Tobias,et al."Comparative validation of multi-instance instrument segmentation in endoscopy: Results of the ROBUST-MIS 2019 challenge".MEDICAL IMAGE ANALYSIS 70(2021):26.