Validating polyp and instrument segmentation methods in colonoscopy through Medico 2020 and MedAI 2021 Challenges.

Autor: Jha D; Machine & Hybrid Intelligence Lab, Department of Radiology, Northwestern University, Chicago, USA. Electronic address: debesh.jha@northwestern.edu., Sharma V; Indian Institute of Technology, Guwahati, India., Banik D; Jadavpur University, Kolkata, India., Bhattacharya D; Institute of Medical Technology and Intelligent Systems, Technische Universität Hamburg, Germany., Roy K; Jadavpur University, Kolkata, India., Hicks SA; SimulaMet, Oslo, Norway., Tomar NK; Machine & Hybrid Intelligence Lab, Department of Radiology, Northwestern University, Chicago, USA., Thambawita V; SimulaMet, Oslo, Norway., Krenzer A; Julius-Maximilian University of Würzburg, Germany., Ji GP; College of Engineering, Australian National University, Canberra, Australia., Poudel S; Department of IT Convergence Engineering, Gachon University, Seongnam 13120, South Korea., Batchkala G; Department of Engineering Science, University of Oxford, Oxford, UK., Alam S; University of Bergen, Bergen, Norway., Ahmed AMA; University of Oslo, Norway., Trinh QH; Faculty of Information Technology, University of Science, VNU-HCM, Viet Nam., Khan Z; National University of Computer and Emerging Sciences, Karachi Campus, Pakistan., Nguyen TP; Faculty of Information Technology, University of Science, VNU-HCM, Viet Nam., Shrestha S; NepAL Applied Mathematics and Informatics Institute for Research (NAAMII), Kathmandu, Nepal., Nathan S; Couger Inc, Tokyo, Japan., Gwak J; Department of Software, Korea National University of Transportation, Chungju-si, South Korea., Jha RK; Machine & Hybrid Intelligence Lab, Department of Radiology, Northwestern University, Chicago, USA., Zhang Z; Machine & Hybrid Intelligence Lab, Department of Radiology, Northwestern University, Chicago, USA., Schlaefer A; Institute of Medical Technology and Intelligent Systems, Technische Universität Hamburg, Germany., Bhattacharjee D; Jadavpur University, Kolkata, India., Bhuyan MK; Indian Institute of Technology, Guwahati, India., Das PK; Indian Institute of Technology, Guwahati, India., Fan DP; Computer Vision Lab (CVL), ETH Zurich, Zurich, Switzerland., Parasa S; Swedish Medical Center, Seattle, USA., Ali S; School of Computing, University of Leeds, LS2 9JT, Leeds, United Kingdom., Riegler MA; SimulaMet, Oslo, Norway; Oslo Metropolitan University, Oslo, Norway. Electronic address: michael@simula.no., Halvorsen P; SimulaMet, Oslo, Norway; Oslo Metropolitan University, Oslo, Norway., de Lange T; Department of Medicine and Emergencies - Mölndal Sahlgrenska University Hospital, Region Västra Götaland, Sweden; Department of Molecular and Clinical Medicin, Sahlgrenska Academy, University of Gothenburg, Sweden., Bagci U; Machine & Hybrid Intelligence Lab, Department of Radiology, Northwestern University, Chicago, USA.
Jazyk: angličtina
Zdroj: Medical image analysis [Med Image Anal] 2024 Sep 05; Vol. 99, pp. 103307. Date of Electronic Publication: 2024 Sep 05.
DOI: 10.1016/j.media.2024.103307
Abstrakt: Automatic analysis of colonoscopy images has been an active field of research motivated by the importance of early detection of precancerous polyps. However, detecting polyps during the live examination can be challenging due to various factors such as variation of skills and experience among the endoscopists, lack of attentiveness, and fatigue leading to a high polyp miss-rate. Therefore, there is a need for an automated system that can flag missed polyps during the examination and improve patient care. Deep learning has emerged as a promising solution to this challenge as it can assist endoscopists in detecting and classifying overlooked polyps and abnormalities in real time, improving the accuracy of diagnosis and enhancing treatment. In addition to the algorithm's accuracy, transparency and interpretability are crucial to explaining the whys and hows of the algorithm's prediction. Further, conclusions based on incorrect decisions may be fatal, especially in medicine. Despite these pitfalls, most algorithms are developed in private data, closed source, or proprietary software, and methods lack reproducibility. Therefore, to promote the development of efficient and transparent methods, we have organized the "Medico automatic polyp segmentation (Medico 2020)" and "MedAI: Transparency in Medical Image Segmentation (MedAI 2021)" competitions. The Medico 2020 challenge received submissions from 17 teams, while the MedAI 2021 challenge also gathered submissions from another 17 distinct teams in the following year. We present a comprehensive summary and analyze each contribution, highlight the strength of the best-performing methods, and discuss the possibility of clinical translations of such methods into the clinic. Our analysis revealed that the participants improved dice coefficient metrics from 0.8607 in 2020 to 0.8993 in 2021 despite adding diverse and challenging frames (containing irregular, smaller, sessile, or flat polyps), which are frequently missed during a routine clinical examination. For the instrument segmentation task, the best team obtained a mean Intersection over union metric of 0.9364. For the transparency task, a multi-disciplinary team, including expert gastroenterologists, accessed each submission and evaluated the team based on open-source practices, failure case analysis, ablation studies, usability and understandability of evaluations to gain a deeper understanding of the models' credibility for clinical deployment. The best team obtained a final transparency score of 21 out of 25. Through the comprehensive analysis of the challenge, we not only highlight the advancements in polyp and surgical instrument segmentation but also encourage subjective evaluation for building more transparent and understandable AI-based colonoscopy systems. Moreover, we discuss the need for multi-center and out-of-distribution testing to address the current limitations of the methods to reduce the cancer burden and improve patient care.
Competing Interests: Declaration of competing interest 1. Financial Interests: Author have no financial interests, direct or indirect, in the research or its outcomes presented in the manuscript. 2. Non-Financial Interests: Author have no non-financial interests that could be perceived as having influenced the research or its presentation in the manuscript. 3. Conflicts of Interest: Author confirm that there are no known conflicts of interest that could potentially bias the results, analysis, or conclusions presented in the manuscript.
(Copyright © 2024 The Author(s). Published by Elsevier B.V. All rights reserved.)
Databáze: MEDLINE
Externí odkaz: