Using Apriori Data Mining Method in COVID-19 Diagnosis

Abstract

Corona virus 2019 (COVID-19) disease has spread all over the world and many people have died due to this disease. PCR (Polymerase Chain Reaction) tests are mostly applied to detect people who have this disease. However, in some cases, it is necessary to wait twenty-four hours for the results of this test. In such cases, the treatment and isolation process of the patient may be delayed. Therefore, the rapid commencement of treatment and isolation process by analyzing the symptoms, are of great importance. Using data mining methods can be carried out quickly specify analysis. Association rule algorithms are also among data mining methods. The most common SETM, AIS and Apriori association rule algorithms are encountered. The most widely used is the Apriori association algorithm. Using this algorithm, the frequency and association rates of the data are found in the data set. In this study, it has been shown that association rules calculated by Apriori algorithm can be used in the diagnosis of COVID-19. By using the COVID-19 Survilance data set, the association rates of the disease symptoms specified in the ICD (International Classification of Diseases) International Classification of Diseases codes were determined. According to the results obtained; it has been observed that the patients with these symptoms are 100% definitely infected with COVID-19 disease when the disease symptoms represented by the A01, A02 and A04 disease codes are together.