[论文解读] Medical Imaging with Deep Learning for COVID- 19 Diagnosis: A Comprehensive Review

Subrato Bharati, Prajoy Podder+2 more|arXiv (Cornell University)|Jul 13, 2021

COVID-19 diagnosis using AI参考文献 145被引用 33

一句话总结

对深度学习在COVID-19医学影像（X-ray和CT）与药物发现中的应用进行全面综述，强调主要DL方法及其性能，以及未来方向。

ABSTRACT

The outbreak of novel coronavirus disease (COVID- 19) has claimed millions of lives and has affected all aspects of human life. This paper focuses on the application of deep learning (DL) models to medical imaging and drug discovery for managing COVID-19 disease. In this article, we detail various medical imaging-based studies such as X-rays and computed tomography (CT) images along with DL methods for classifying COVID-19 affected versus pneumonia. The applications of DL techniques to medical images are further described in terms of image localization, segmentation, registration, and classification leading to COVID-19 detection. The reviews of recent papers indicate that the highest classification accuracy of 99.80% is obtained when InstaCovNet-19 DL method is applied to an X-ray dataset of 361 COVID-19 patients, 362 pneumonia patients and 365 normal people. Furthermore, it can be seen that the best classification accuracy of 99.054% can be achieved when EDL_COVID DL method is applied to a CT image dataset of 7500 samples where COVID-19 patients, lung tumor patients and normal people are equal in number. Moreover, we illustrate the potential DL techniques in drug or vaccine discovery in combating the coronavirus. Finally, we address a number of problems, concerns and future research directions relevant to DL applications for COVID-19.

研究动机与目标

• 总结 DL 模型在 COVID-19 检测的医学影像中的应用（X-ray、CT）以及比较性能。
• 解释在 COVID-19 影像中超越分类的 DL 任务（定位、分割、配准）。
• 讨论 DL 在药物/疫苗发现中的应用并指出当前挑战与未来研究方向。

提出的方法

• 对 COVID-19 成像中图像分类、定位、分割和配准的 DL 方法进行综述。
• 整理并比较在 X-ray 和 CT 数据集上报告的准确性指标。
• 综合用于 COVID-19 药物与疫苗发现的 DL 技术。

实验结果

研究问题

• RQ1 哪些 DL 架构和预处理策略能在 X-ray 和 CT 图像中实现 COVID-19 检测的最高准确性？
• RQ2 在 COVID-19 图像分类及相关任务（定位、分割、配准）中表现最好的 DL 方法有哪些？
• RQ3 在支持 COVID-19 药物/疫苗发现方面，潜在的 DL 方法与挑战有哪些？

主要发现

• 在 X-ray 分类中报道的最高准确率为 99.80%，使用 InstaCovNet-19，在一个包含 361 例 COVID-19、362 例肺炎和 365 例正常病例的数据集上。
• 在 CT 分类中报道的最高准确率为 99.054%，使用 EDL_COVID，在一个7500样本的数据集中，COVID-19、肺肿瘤、正常病例数量相等。
• DL 技术在 COVID-19 成像中不仅限于分类，还扩展到定位、分割和配准。
• 讨论了 DL 在 COVID-19 的药物或疫苗发现中的潜在影响领域。
• 本文亦探讨了在 COVID-19 的 DL 领域中的问题、担忧与未来研究方向。