Reference: http://ocel.ai/story-telling/
- Who are the people or communities in need of help? This would help any patient suffering with viral and bacterial infections who are experiencing symptoms like fever, cough, fatigue, headaches and shortness of breath.
- What problem happened to them? Patients with COVID-19 infection need to be diagnosed quickly and accurately in order to successfully treat them. Currently, COVID -19 can be detected by RT-PCR however, it is limited and it has a higher false positive and false negative results than the desired rates. Therefore, to assist the traditional methodology for accurate clinical diagnosis, and to estimate the severity of the respiratory disease COVID-19 can be screened using X-Ray images of the lungs of an individual. However, it is hard to diagnose COVID-19 from pneumonia caused by other viral and bacterial infections based only on the clinical judgement of chest radiograph images in a short period of time especially in the absence of a radiologist.Detecting COVID-19 patients accurately using X-Ray images in an automated manner through modern AI techniques could greatly alleviate this problem.
- When did the problem take place? Where means two things: 1) The environment and settings that the people or the community is living in, and 2) the place/location where the problem take place? According to WHO report the first case of COVID was reported from Wuhan, China and was spread globally in just a few months. However, there has been no clear evidence what caused this virus to spread rapidly. Some research indicates this has started from Wuhan wet market and some indicate it has spread from Bats infestation. COVID shows similar symptoms as Pneumonia like fluid in the lungs, suffocations, fever and shortness of breath. However, when compared deeply there seems to be a great variation in the X-rays and other additional symptoms. This application will help people to analyse x-rays effectively and can determine the difference between COVID case and regular Pneumonia case.
- Why means the possible causes and/or origin of the problem? Pneumonia is caused by bacterial and viral infection but most commonly by a bacteria called Streptococcus pneumonia. Pneumonia can also be caused by a coronavirus called SARS-CoV-2 which is termed as COVID-19. The rapidly increasing COVID-19 in patients is overwhelming healthcare systems across the world. The limited conventional testing kits (RT-PCR) have long turn-around time and limited sensitivity making it impossible for every patient with COVID symptoms to be tested. Detecting COVID-19 infections using Chest X-Ray may help identify the high risk patients quickly as X-Ray machines are already available in most healthcare systems. Identifying pneumonia through radiography images may be readily possible. However, detecting COVID-19 infections among pneumonia caused by bacteria and other viral organisms is hard to differentiate.
- How: If you would like, you can add a dimension of how. How did it happen? Sometimes, the answer to how can be covered by what, when and where? This project tackles a particular medical scenario where the COVID infected patient is diagnosed with Pneumonia. The idea of predicting COVID from Chest X-Ray images would require highly skilled and trained Radiologists in our disposal which is highly ambitious. Also, reading numerous Chest X-Rays is time consuming even if it has been done by trained professionals. Developing an AI model would diminish this problem.
