Until now, Coronavirus SARS-CoV-2 has caused more than 850,000 deaths and infected more than 27 million individuals in over 120 countries. Besides principal polymerase chain reaction (PCR) tests, automatically identifying positive samples based on computed tomography (CT) scans can present a promising option in the early diagnosis of COVID-19. Recently, there have been increasing efforts to utilize deep networks for COVID-19 diagnosis based on CT scans. While these approaches mostly focus on introducing novel architectures, transfer learning techniques, or construction large scale data, we propose a novel strategy to improve the performance of several baselines by leveraging multiple useful information sources relevant to doctors' judgments. Specifically, infected regions and heat maps extracted from learned networks are integrated with the global image via an attention mechanism during the learning process. This procedure not only makes our system more robust to noise but also guides the network focusing on local lesion areas. Extensive experiments illustrate the superior performance of our approach compared to recent baselines. Furthermore, our learned network guidance presents an explainable feature to doctors as we can understand the connection between input and output in a grey-box model.

中文翻译：

---

从 CT 图像中检测 COVID-19 的具有多个知识源的注意力机制

迄今为止，冠状病毒 SARS-CoV-2 已在 120 多个国家/地区造成超过 850,000 人死亡并感染了超过 2700 万人。除了主要的聚合酶链反应 (PCR) 测试之外，基于计算机断层扫描 (CT) 扫描自动识别阳性样本可以为 COVID-19 的早期诊断提供一个有希望的选择。最近，越来越多的努力利用深度网络进行基于 CT 扫描的 COVID-19 诊断。虽然这些方法主要侧重于引入新颖的架构、迁移学习技术或构建大规模数据，但我们提出了一种新颖的策略，通过利用与医生判断相关的多个有用信息源来提高多个基线的性能。具体来说，从学习网络中提取的受感染区域和热图在学习过程中通过注意力机制与全局图像集成。这个过程不仅使我们的系统对噪声更加鲁棒，而且还引导网络关注局部病变区域。大量实验表明，与最近的基线相比，我们的方法具有优越的性能。此外，我们学习到的网络指导为医生提供了一个可解释的特征，因为我们可以理解灰盒模型中输入和输出之间的联系。