GraphXCOVID: Explainable deep graph diffusion pseudo-Labelling for identifying COVID-19 on chest X-rays.
GraphXCOVID: Explainable deep graph diffusion pseudo-Labelling for identifying COVID-19 on chest X-rays.
复制标题
DOI：
10.1016/j.patcog.2021.108274
复制
发表时间：
2022-03
影响因子：
8
通讯作者：
Papadakis N
中科院分区：
文献类型：
--
作者：
Aviles-Rivero AI;Sellars P;Schönlieb CB;Papadakis N
一个人能在最低限度的监督下学会诊断COVID-19吗？自新型COVID-19爆发以来，人们纷纷开发自动技术，用于对胸部X射线数据进行专家级疾病识别。特别是，深度监督学习的使用已成为首选范例。然而，这种模型的性能在很大程度上取决于一个大的和有代表性的标记数据集的可用性。其创建是一项非常昂贵和耗时的任务，尤其是对新疾病提出了巨大挑战。半监督学习已经显示出能够匹配监督模型的令人难以置信的性能，同时需要一小部分标记的示例。这使得半监督模式成为识别COVID-19的有吸引力的选择。在这项工作中，我们引入了一个基于图的深度半监督框架，用于从胸部X射线中分类COVID-19。我们的框架引入了一个优化模型的图形扩散，加强了微小的标记集和巨大的未标记的数据之间的自然关系。然后，我们将扩散预测输出连接为伪标签，用于深度网络中的迭代方案。我们通过实验证明，我们的模型能够用一小部分标记的例子胜过当前领先的监督模型。最后，我们提供了注意力地图，以适应放射科医生的心理模型，更好地适应他们的感知和认知能力。这些可视化旨在帮助放射科医生判断诊断是否正确，从而加快决策。
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.
登录
查看更多内容
影响因子：
2
作者：
Apostolopoulos, Ioannis D.;Aznaouridis, Sokratis I.;Tzani, Mpesiana A.
通讯作者：
Tzani, Mpesiana A.
影响因子：
19.7
作者：
Fang, Yicheng;Zhang, Huangqi;Ji, Wenbin
通讯作者：
Ji, Wenbin
影响因子：
10.6
作者：
Chen, Hanbo;Li, Kaiming;Liu, Tianming
通讯作者：
Liu, Tianming
影响因子：
5.9
作者：
Chung, Mi Sun;Lee, Eun Jung;Byun, Jun Soo
通讯作者：
Byun, Jun Soo
影响因子：
4.6
作者：
Baltruschat, Ivo M.;Nickisch, Hannes;Saalbach, Axel
通讯作者：
Saalbach, Axel