Endoscopy is a widely used clinical procedure for the early detection of numerous cancers (e.g., nasopharyngeal, oesophageal adenocarcinoma, gastric, colorectal and bladder cancers) therapeutic procedures and minimally invasive surgery (e.g., laparoscopy). A major drawback of endoscopy video frames is that they are heavily corrupted with multiple imaging artefacts (e.g., pixel saturations, motion blur, defocus, specular re ections, bubbles, uid and debris). These artefacts not only present di culty in visualizing the underlying tissue during diagnosis but also a ect any post-analysis methods such as lesion detection and video image retrieval required for informative reporting and follow-ups. Accurate detection of artefacts is therefore a core challenge in a wide-range of endoscopic applications across multiple disease areas. The precise detection of endoscopic artefacts is crucial to achieve high-quality endoscopic frame restoration and necessary to realize reliable computer assisted endoscopy tools for improved patient care.
Together with six international clinical data providers, we collated a large set of multi-organ, multi-modal and multi-population endoscopy video frames corrupted with di erent artefacts. Gold standard ground truths were established and made publicly available to the community. This challenge comprised of three sub-challenges: 1) detection, 2) semantic segmentation and 3) detection generalization. All algorithms were evaluated online with the same evaluation metrics. We attracted nearly 29 AI algorithm developers and 529 enthusiasts to participate in the EAD2019 challenge.
This volume contains the proceedings of the rst edition of the EAD challenge workshop, which was held in Venice, Italy, on April 8, 2019 co-located with the 16th International Symposium on Biomedical Imaging (ISBI). This editiion of the EAD workshop has been compiled from the 10 accepted papers submitted by top performing teams (12 out of 29) in an endoscopy artefact detection challenge1. Each paper was reviewed by at least 2 reviewers in the eld. Workshop and Challenge Organizers: Sharib Ali, Felix Zhou and Christian Daul 1 https://ead2019.grand-challenge.org
EAD2019 Challenge Organization
Department of Engineering Science, University of Oxford, Oxford, UK Welcome Trust Centre for Human Genetics, Oxford, UK University of Lorraine, CNRS, CRAN, UMR 7039, Nancy, France Department of Information Engineering, University of Padova, Padova, Italy EPFL, Lausanne, Switzerland Prokhorov General Physics Institute of the Russian Academy of Sciences, Moscow, Russia Johns Hopkins University, Baltimore, USA
Translational Gastroenterology Unit, John Radcli e Hospital, Oxford, UK Translational Gastroenterology Unit, John Radcli e Hospital, Oxford, UK Translational Gastroenterology Unit, John Radcli e Hospital, Oxford, UK Istituto Oncologico Veneto, IOV-IRCCS, Padova, Italy Medical Image Analysis Network (MedIAN), UK Cancer Research UK (CR UK) Oxford, UK
Department of Engineering Science, University of Oxford, Oxford, UK University of Lorraine, CNRS, CRAN, UMR 7039, Nancy, France IBME, BDI, Department of Engineering Science, University of Oxford, Ofxord, UK Ludwig Cancer Institute, University of Oxford, Oxford, UK Universite Clermont Auvergne, ClermontFerrand, France Translational Gastroenterology Unit, John Radcli e Hospital, Oxford, UK Translational Gastroenterology Unit, John Radcli e Hospital, Oxford, UK
Khanal, Bishesh