Noran Hany, Nura Mostafa, Nourhan Atef, Sara Mohamed

Publishing Date

October 2020


The main idea of this project is to detect COVID-19 and pneumonia and discriminate between them. COVID-19 is an infectious disease that mainly affects lungs. Pneumonia is also an infection that causes inflammation in the air sacs in lungs. COVID-19 and pneumonia have similar symptoms in lungs as breathing problems that appear in CT as ground-glass opacity (GGO). For discrimination between them, we are using a dataset that consists of CT images of three classes (COVID-19, Pneumonia, Normal lungs), for determining the severity of each of them, we are using K-means clustering. And for COVID-19 detection we will use a dataset which contain 6 blood tests and PCR for COVID-19 and NON-COVID-19 patients. By using deep learning techniques, which will facilitate diagnosing these diseases efficiently and in a shorter time, it is expected to provide better results compared to other techniques. The Convolutional Neural Networks model (CNN) is proposed to be used in differentiating between these two diseases and get better results. It is considered a high sensitivity model to characterize and diagnose COVID-19 infections. 

1.1 Background

Coronavirus is an infectious disease that spread around the world. Most of the people that got infected with the COVID-19 virus suffered from respiration problems. Older people who suffer from medical problems like heart disease, diabetes, respiratory disease and cancer have a poor prognosis. It spreads among people by having close interactions with others who are infected. PCR test is the current solution for detecting these diseases, but it is not the preferred solution. IT is not accurate as CT scan images in addition to that CT scan offers a faster solution over PCR. Pneumonia also is a respiratory disease that many people suffer from around the world. The total death has been recording more than one million children a year. Therefore, our system aims to detect both of diseases and differentiate between them through CT scan images and determine the severity. Deep learning techniques will facilitate and help in detecting and discriminating between COVID-19 and pneumonia by using CNN algorithms.

1.2 Motivation

Academic: COVID-19 and pneumonia are respiratory diseases. There are more than 43 Million person infected by COVID-19. And a lot of deaths from pneumonia have been recorded around the world from 1990 to 2017. Researchers are trying to find a method to detectCOVID-19 early and before reaching its final stages and discriminate between it and pneumonia. But they didn’t reach a successful solution, because most of them used x-ray images in detection. X-ray images don’t give many details about the lungs. Li et al., They included that CT images are more accurate and detect COVID-19 and types of pneumonia than PCR. Hu et al., They included that PCR sensitivity is not high enough to detect COVID-19. So they used CT scan images. Qjidaa et al.,They used x-ray images to solve the PCR problem. They used deep learning with CNN models to detect COVID-19 and discriminate between it and pneumonia.

Business: COVID-19 is spreading around the world. It caused a major economic shock. In this paper, we explore the impact of COVID-19 on the business and the economy around the world, showing its utterly devastating effects and how the system would help to overcome these crucial problems before they happen. Another major advantage is that this system can save time and effort, especially during the pandemic, as it will be able to detect the disease with the system by using CT images. This will lead to fewer patients visiting hospitals in these hard times. The economy got affected a lot because of COVID-19, as the world was moving towards economic distress, and still, the economy has not been fully recovered from this economic disaster. So basically, pandemics have a direct effect on countries’ gross domestic profit which directly affects the economy, happened to the whole world as well as here in Egypt. The graph of the stock exchange in Egypt EGX30 which represents the top 30Large companies that are leading the stock exchange in Egypt, the coronavirus was widely spread in Egypt from February-March 2020 until now. So, it’s obvious from the graph how it got too steep at this period, and still, it didn’t recover until this moment. Moreover, moving to the second point which is large businesses, because of COVID-19 many companies had to cut off the number of employees and restructure the company’s hierarchy, increasing the unemployment percentage.

1.3 Problem Statement

Doctors face difficulties diagnosing COVID-19 and pneumonia because of the similarity between them. Also, the majority of people do a PCR test to know if they have COVID-19 or not. PCR compared to CT scan images is more expensive, takes more time, and is not as accurate as CT. Also the sensitivity of PCR is 83.3% while the sensitivity of CT is 97.2 %.