Authors: Chris P. Tsokos, Lohuwa Mamudu.
To address the testing of the horrific pandemic disease that has terrified our global society, COVID-19, we have developed an analytical model that an individual can easily apply to determine if he or she tested positive or negative with a very high degree of accuracy. Our analytical model is real data-driven utilizing data obtained from the World Health Organization, WHO, and the United States Center for Disease Control and Prevention, CDC guidelines. Both WHO and CDC have identified several symptoms or risk factors from individuals diagnosed with the disease, COVID-19. They have identified and published nine symptoms that are associated with the disease, COVID-19. However, our structured analytical model identified only seven of the nine symptoms to statistically significantly contribute to the subject disease. They are fever, tiredness, dry cough, difficulty in breathing, sore throat, pain, and nasal congestion. Each of the symptoms shows highly likelihood of having COVID-19. Our analytical model was carefully developed, very well-validated, and statistically tested to achieve a 93% accuracy in the testing result. If a person is tested positive, we recommend that he/she seek medical evaluation and treatment. That is, once we receive the categorical data from a given individual, and we input into the proposed model, the output result will be the individual is tested positive or negative for COVID-19. The developed model identifies (estimated) the different weights of each of the seven symptoms or risk factors that play a major role in the decision process of the testing results. Our findings seek to enhance testing efficiency, treatment, control, and prevention strategy for the COVID-19 disease.
View/Download pdf