Background After the spread of COVID-19 pneumonia, chest CT examination was used as a substantial non-invasive complement to RT-PCR for diagnosing and as a rapid screening tool when RT-PCR results are unavailable. Our study aimed at the analysis of the lung abnormalities detected by chest CT in COVID-19 pneumonia according to the severity and duration of symptoms. Results In the early phase (n = 60), 32 patients had negative CT findings and 28 patients had positive findings with a mean total lung severity score of 2.13. In the intermediate phase (n = 116), 4 patients had negative CT findings and 112 patients had positive findings with a mean total lung severity score of 16.08. In the late phase (n = 36), all patients had positive findings with a mean total lung severity score of 17.83. CT lung abnormalities were progressed on follow-up CT studies. We found a high total lung severity score in many patients with mild symptoms with a mean of 14.77 and a low total lung severity score in many patients with moderate to severe symptoms with a mean of 9.14. Conclusion Chest CT should be used as a routine examination for diagnosing COVID-19 pneumonia and follow-up of disease advance. The progression of lung abnormalities was related to the duration more than the severity of symptoms.

<span id="docs-internal-guid-cb130a3a-7fff-3e11-ae3d-ad2310e265f8"><span>Deep learning (DL) algorithms achieved state-of-the-art performance in computer vision, speech recognition, and natural language processing (NLP). In this paper, we enhance the convolutional neural network (CNN) algorithm to classify cancer articles according to cancer hallmarks. The model implements a recent word embedding technique in the embedding layer. This technique uses the concept of distributed phrase representation and multi-word phrases embedding. The proposed model enhances the performance of the existing model used for biomedical text classification. The result of the proposed model overcomes the previous model by achieving an F-score equal to 83.87% using an unsupervised technique that trained on PubMed abstracts called PMC vectors (PMCVec) embedding. Also, we made another experiment on the same dataset using the recurrent neural network (RNN) algorithm with two different word embeddings Google news and PMCVec which achieving F-score equal to 74.9% and 76.26%, respectively.</span></span>

No abstract