Abstract
According to the published reports and studies, the symptoms of the disease caused by the COVID-19 virus have not yet been fully determined. It is a major stress on clinicians to make a correct and consistent decision about whether to apply the test or not, as many factors with extreme uncertainty need to be evaluated at once. In this study, it is aimed to provide assistance to the clinicians by processing the data using fuzzy logic based decision support system at the time of the decision-making process. In the designed fuzzy logic based decision support system, a fuzzy rule-base was created with linguistic information by interpreting the symptoms that are naturally uncertain by experts. With the help of the obtained fuzzy rule base, the input data of symptoms will be processed and the risk of a person being infected will be obtained as an output. As the results of the estimation module constructed with the existing parameters are examined, it is observed to be compatible with the data published before. In this context, a data set with 50 different patients were designed randomly to evaluate the system. For the analysis of the nonlinear mapping obtained with the Mamdani type fuzzy inference system, random test data is used and infection risk at rates varying between 12.5-83% was determined. The fuzzy logic based decision support system for COVID-19 can be accepted as applicable, flexible, and trustworthy for clinicians. It can be said that this system is not only suitable for COVID-19 but also applicable for future epidemics.
Highlights
Disease is often defined as any impairment, disorder or disability in the functions or physiological, biological structure of the human body (Hunter, 2009)
It is difficult to determine the nonlinear nature of the symptom, disease, and diagnostic process consisting of uncertainties with the two-valued classical logic
The literature published in the determination of the most common symptoms of infected patients, as well as the reports of World Health Organization (WHO), Republic of Turkey Ministry of Health and other countries have been taken as a source (RTMH, 2021; WHO, 2021)
Summary
Disease is often defined as any impairment, disorder or disability in the functions or physiological, biological structure of the human body (Hunter, 2009). It is diagnosed by physician’s interpretation of characteristic signs and symptoms. There are no clear standard case definitions for the diagnosis of a disease. Due to these uncertain signs and symptoms, the decision-making process for a physician becomes more complex. Computer-aided systems have been used to assist clinicians in this complex and uncertain disease diagnosis process. Fuzzy logic as a class of AI method has been used to model the aforementioned complexities of the decision process
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
