Integration of CNN, CBMIR, and Visualization Techniques for Diagnosis and Quantification of Covid-19 Disease

• DOI: 10.1109/jbhi.2021.3067333
• Publication Date: 2021-03-18T19:24:39Z
• Authors: Saeed Mohagheghi; Mehdi Alizadeh; Seyed Mahdi Safavi; Amir H. Foruzan; Yen-Wei Chen
• Keywords: Transfer of learning

Diagnosis techniques based on medical image modalities have higher sensitivities compared to conventional RT-PCT tests. We propose two methods for diagnosing COVID-19 disease using X-ray images and differentiating it from viral pneumonia. The diagnosis section is deep neural networks, the discriminating uses an retrieval approach. Both units were trained by healthy, pneumonia, images. In patients, maximum intensity projection of lung CT visualized a physician, Involvement Score calculated. performance CNN algorithms improved transfer learning hashing functions. achieved accuracy 97% overall prec@10 87%, respectively, concerning methods.