Sarp Salih, Catak Ferhat Ozgur, Kuzlu Murat, Cali Umit, Kusetogullari Huseyin, Zhao Yanxiao, Ates Gungor, Guler Ozgur
Electrical & Computer Engineering, Virginia Commonwealth University, Richmond, VA, USA.
Department of Electrical Engineering & Computer Science, University of Stavanger, Rogaland, Norway.
Heliyon. 2023 Apr;9(4):e15137. doi: 10.1016/j.heliyon.2023.e15137. Epub 2023 Apr 7.
The coronavirus disease (COVID-19) has continued to cause severe challenges during this unprecedented time, affecting every part of daily life in terms of health, economics, and social development. There is an increasing demand for chest X-ray (CXR) scans, as pneumonia is the primary and vital complication of COVID-19. CXR is widely used as a screening tool for lung-related diseases due to its simple and relatively inexpensive application. However, these scans require expert radiologists to interpret the results for clinical decisions, i.e., diagnosis, treatment, and prognosis. The digitalization of various sectors, including healthcare, has accelerated during the pandemic, with the use and importance of Artificial Intelligence (AI) dramatically increasing. This paper proposes a model using an Explainable Artificial Intelligence (XAI) technique to detect and interpret COVID-19 positive CXR images. We further analyze the impact of COVID-19 positive CXR images using heatmaps. The proposed model leverages transfer learning and data augmentation techniques for faster and more adequate model training. Lung segmentation is applied to enhance the model performance further. We conducted a pre-trained network comparison with the highest classification performance (F1-Score: 98%) using the ResNet model.
在这个前所未有的时期，冠状病毒病（COVID-19）持续带来严峻挑战，在健康、经济和社会发展等方面影响着日常生活的方方面面。由于肺炎是COVID-19的主要且关键的并发症，对胸部X线（CXR）扫描的需求日益增加。CXR因其应用简单且成本相对较低，被广泛用作肺部相关疾病的筛查工具。然而，这些扫描需要专业放射科医生解读结果以做出临床决策，即诊断、治疗和预后判断。在疫情期间，包括医疗保健在内的各个领域的数字化进程加速，人工智能（AI）的应用和重要性大幅提升。本文提出一种使用可解释人工智能（XAI）技术的模型，用于检测和解读COVID-19阳性的CXR图像。我们还使用热图进一步分析COVID-19阳性CXR图像的影响。所提出的模型利用迁移学习和数据增强技术，以实现更快且更充分的模型训练。应用肺部分割技术进一步提升模型性能。我们使用ResNet模型与具有最高分类性能（F1分数：98%）的预训练网络进行了比较。
