Human-AI collaborative multi-modal multi-rater learning for endometriosis diagnosis.

Autor:	Wang H; Computer Vision, Mohamed bin Zayed University of Artificial Intelligence, Mohamed bin Zayed University of Artificial Intelligence (MBZUAI), Masdar City, Abu Dhabi, UAE, Masdar City, Abu Dhabi, UNITED ARAB EMIRATES., Butler D; Computer Science, The University of Adelaide, The University of Adelaide, Adelaide, South Australia 5005, Adelaide, South Australia, 5005, AUSTRALIA., Zhang Y; Computer Science, The University of Adelaide, School of Computer Science, The University of Adelaide, Adelaide, Adelaide, South Australia, 5005, AUSTRALIA., Avery J; Robinson Research Institute, The University of Adelaide, Robinson Research Institute, The University of Adelaide, Adelaide, Adelaide, South Australia, 5005, AUSTRALIA., Knox S; Benson Radiology, Benson Radiology, Adelaide, Wayville, South Australia, 5034, AUSTRALIA., Ma C; Macquarie University, Macquarie University, Balaclava Rd, Macquarie Park NSW 2109, Australia, Sydney, New South Wales, 2109, AUSTRALIA., Hull L; Robinson Research Institute, The University of Adelaide, Robinson Research Institute, The University of Adelaide, Adelaide, Adelaide, South Australia, 5005, AUSTRALIA., Carneiro G; University of Surrey, University of Surrey, UK, Guildford, Surrey, GU2 7XH, UNITED KINGDOM OF GREAT BRITAIN AND NORTHERN IRELAND.
Jazyk:	angličtina
Zdroj:	Physics in medicine and biology [Phys Med Biol] 2024 Dec 02. Date of Electronic Publication: 2024 Dec 02.
DOI:	10.1088/1361-6560/ad997e
Abstrakt:	Endometriosis, affecting about 10% of individuals assigned female at birth, is challenging to diagnose and manage. Diagnosis typically involves the identification of various signs of the disease using either laparoscopic surgery or the analysis of T1/T2 MRI images, with the latter being quicker and cheaper but less accurate. A key diagnostic sign of endometriosis is the obliteration of the Pouch of Douglas (POD). However, even experienced clinicians struggle with accurately classifying POD obliteration from MRI images, which complicates the training of reliable AI models. In this paper, we introduce the Human-AI Collaborative Multi-modal Multi-rater Learning (HAICOMM) methodology to address the challenge above. HAICOMM is the first method that explores three important aspects of this problem: 1) multi-rater learning to extract a cleaner label from the multiple "noisy" labels available per training sample; 2) multi-modal learning to leverage the presence of T1/T2 MRI images for training and testing; and 3) human-AI collaboration to build a system that leverages the predictions from clinicians and the AI model to provide more accurate classification than standalone clinicians and AI models. Presenting results on the multi-rater T1/T2 MRI endometriosis dataset that we collected to validate our methodology, the proposed HAICOMM model outperforms an ensemble of clinicians, noisy-label learning models, and multi-rater learning methods. (© 2024 Institute of Physics and Engineering in Medicine. All rights, including for text and data mining, AI training, and similar technologies, are reserved.)
Databáze:	MEDLINE
Externí odkaz:	Zobrazit plný text záznamu