期刊
PATTERN RECOGNITION
卷 122, 期 -, 页码 -出版社
ELSEVIER SCI LTD
DOI: 10.1016/j.patcog.2021.108274
关键词
COVID-19; Chest X-ray; Semi-Supervised learning; Deep learning; Explainability
资金
- University of Cambridge
- EPSRC [EP/S026045/1, EP/T003553/1]
- NPL
- Leverhulme Trust
- Philip Leverhulme Prize
- Royal Society Wolfson Fellowship
- EPSRC Centre [EP/N014588/1]
- European Union [777826]
- Alan Turing Institute
- CMIH
- CCIMI
- Wellcome Innovator Award [RG98755]
- CHiPS
Semi-supervised learning shows promise in COVID-19 diagnosis, outperforming supervised models with just a small fraction of labeled examples. Researchers introduce a graph-based deep semi-supervised framework that optimizes the relation between labeled and unlabeled data, and provide visualizations to assist radiologists in diagnosis.
Can one learn to diagnose COVID-19 under extreme minimal supervision? Since the outbreak of the novel COVID-19 there has been a rush for developing automatic techniques for expert-level disease identification on Chest X-ray data. In particular, the use of deep supervised learning has become the go-to paradigm. However, the performance of such models is heavily dependent on the availability of a large and representative labelled dataset. The creation of which is a heavily expensive and time consuming task, and especially imposes a great challenge for a novel disease. Semi-supervised learning has shown the ability to match the incredible performance of supervised models whilst requiring a small fraction of the labelled examples. This makes the semi supervised paradigm an attractive option for identifying COVID-19. In this work, we introduce a graph based deep semi-supervised framework for classifying COVID-19 from chest X-rays. Our framework introduces an optimisation model for graph diffusion that reinforces the natural relation among the tiny labelled set and the vast unlabelled data. We then connect the diffusion prediction output as pseudo-labels that are used in an iterative scheme in a deep net. We demonstrate, through our experiments, that our model is able to outperform the current leading supervised model with a tiny fraction of the labelled examples. Finally, we provide attention maps to accommodate the radiologist's mental model, better fitting their perceptual and cognitive abilities. These visualisation aims to assist the radiologist in judging whether the diagnostic is correct or not, and in consequence to accelerate the decision. (C) 2021 Elsevier Ltd. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据