Rapid artificial intelligence solutions in a pandemic-The COVID-19-20 Lung CT Lesion Segmentation Challenge

Authors

• Holger R. Roth, NVIDIA, Bethesda, MD, USA; Santa Clara, CA, USA; Munich, Germany. Electronic address: hroth@nvidia.com.
• Ziyue Xu, NVIDIA, Bethesda, MD, USA; Santa Clara, CA, USA; Munich, Germany.
• Carlos Tor-Díez, Sheikh Zayed Institute for Pediatric Surgical Innovation, Children's National Hospital, WA, DC, USA.
• Ramon Sanchez Jacob, Division of Diagnostic Imaging and Radiology, Children's National Hospital, WA,DC, USA.
• Jonathan Zember, Division of Diagnostic Imaging and Radiology, Children's National Hospital, WA,DC, USA.
• Jose Molto, Division of Diagnostic Imaging and Radiology, Children's National Hospital, WA,DC, USA.
• Wenqi Li, NVIDIA, Bethesda, MD, USA; Santa Clara, CA, USA; Munich, Germany.
• Sheng Xu, Radiology and Imaging Sciences, National Institutes of Health Clinical Center, Bethesda, MD, USA.
• Baris Turkbey, Radiology and Imaging Sciences, National Institutes of Health Clinical Center, Bethesda, MD, USA.
• Evrim Turkbey, Radiology and Imaging Sciences, National Institutes of Health Clinical Center, Bethesda, MD, USA.
• Dong Yang, NVIDIA, Bethesda, MD, USA; Santa Clara, CA, USA; Munich, Germany.
• Ahmed Harouni, NVIDIA, Bethesda, MD, USA; Santa Clara, CA, USA; Munich, Germany.
• Nicola Rieke, NVIDIA, Bethesda, MD, USA; Santa Clara, CA, USA; Munich, Germany.
• Shishuai Hu, School of Computer Science and Engineering, Northwestern Polytechnical University, China.
• Fabian Isensee, Applied Computer Vision Lab, Helmholtz Imaging , Heidelberg, Germany; Division of Medical Image Computing, German Cancer Research Center (DKFZ), Heidelberg, Germany.
• Claire Tang, Lynbrook High School, San Jose, CA, USA.
• Qinji Yu, Shanghai Jiao Tong University, China.
• Jan Sölter, Luxembourg Centre for Systems Biomedicine, University of Luxembourg, Luxembourg.
• Tong Zheng, School of Informatics, Nagoya University, Japan.
• Vitali Liauchuk, Biomedical Image Analysis Department, United Institute of Informatics Problems, Belarus.
• Ziqi Zhou, Guangdong Key Laboratory of Intelligent Information Processing, Shenzhen University, China.
• Jan Hendrik Moltz, Fraunhofer Institute for Digital Medicine MEVIS, Bremen, Germany.
• Bruno Oliveira, Life and Health Sciences Research Institute (ICVS), School of Medicine, University of Minho, Braga, Portugal; ICVS/3B's - PT Government Associate Laboratory, Braga/Guimarães, Portugal; Algoritmi Center, School of Engineering, University of Minho, Guimarães, Portugal; 2Ai - School of Technology, IPCA, Barcelos, Portugal.
• Yong Xia, School of Computer Science and Engineering, Northwestern Polytechnical University, China.
• Klaus H. Maier-Hein, Pattern Analysis and Learning Group, Department of Radiation Oncology, Heidelberg University Hospital, Heidelberg, Germany.
• Qikai Li, Shanghai Jiao Tong University, China.
• Andreas Husch, Fraunhofer Institute for Digital Medicine MEVIS, Lübeck, Germany.
• Luyang Zhang, School of Informatics, Nagoya University, Japan.
• Vassili Kovalev, Biomedical Image Analysis Department, United Institute of Informatics Problems, Belarus.
• Li Kang, Guangdong Key Laboratory of Intelligent Information Processing, Shenzhen University, China.
• Alessa Hering, Fraunhofer Institute for Digital Medicine MEVIS, Lübeck, Germany.
• João L. Vilaça, Ai - School of Technology, IPCA, Barcelos, Portugal.

Document Type

Journal Article

Publication Date

9-6-2022

Journal

Medical image analysis

Volume

82

DOI

10.1016/j.media.2022.102605

Keywords

COVID-19; Challenge; Medical image segmentation

Abstract

Artificial intelligence (AI) methods for the automatic detection and quantification of COVID-19 lesions in chest computed tomography (CT) might play an important role in the monitoring and management of the disease. We organized an international challenge and competition for the development and comparison of AI algorithms for this task, which we supported with public data and state-of-the-art benchmark methods. Board Certified Radiologists annotated 295 public images from two sources (A and B) for algorithms training (n=199, source A), validation (n=50, source A) and testing (n=23, source A; n=23, source B). There were 1,096 registered teams of which 225 and 98 completed the validation and testing phases, respectively. The challenge showed that AI models could be rapidly designed by diverse teams with the potential to measure disease or facilitate timely and patient-specific interventions. This paper provides an overview and the major outcomes of the COVID-19 Lung CT Lesion Segmentation Challenge - 2020.

APA Citation

Roth, Holger R.; Xu, Ziyue; Tor-Díez, Carlos; Sanchez Jacob, Ramon; Zember, Jonathan; Molto, Jose; Li, Wenqi; Xu, Sheng; Turkbey, Baris; Turkbey, Evrim; Yang, Dong; Harouni, Ahmed; Rieke, Nicola; Hu, Shishuai; Isensee, Fabian; Tang, Claire; Yu, Qinji; Sölter, Jan; Zheng, Tong; Liauchuk, Vitali; Zhou, Ziqi; Moltz, Jan Hendrik; Oliveira, Bruno; Xia, Yong; Maier-Hein, Klaus H.; Li, Qikai; Husch, Andreas; Zhang, Luyang; Kovalev, Vassili; Kang, Li; Hering, Alessa; and Vilaça, João L., "Rapid artificial intelligence solutions in a pandemic-The COVID-19-20 Lung CT Lesion Segmentation Challenge" (2022). GW Authored Works. Paper 1660.
https://hsrc.himmelfarb.gwu.edu/gwhpubs/1660

Department

Radiology