A Hybrid Decision Support System Application with Analytic Hierarchy Process and Data Mining Techniques: Diagnosis of Covid19 with Complete Blood Count Values
Project no. 23066 · Status: Kapanmış (Yayın bekleniyor) · 2021
Summary
<p>Data mining techniques significantly improve the accuracy of artificial intelligence-based diagnostics. In this study, a feature selection method and methods that overcome the class imbalance problem with data mining techniques were used to the complete blood count (hemogram) values of Covid 19 patients. The analytic hierarchy process (AHP) method, one of the multi-criteria decision-making approaches, was applied to the variables whose number was reduced in the balanced data set. Using the AHP method and expert opinions, the priorities of the variables achieved by machine learning were determined and a decision model was created with publicly available data. A web application of this decision model was designed to deliver the decision support system to end users. In addition, the usability of the decision support system and user satisfaction were evaluated. With this decision support, it will be possible to predict the status of the patient's Covid-19 disease with the results of the hemogram values, until the imaging and PCR test results of a person who comes to the hospital with the suspicion of Covid-19 disease were obtained. As a result, it may be possible to isolate the patient more quickly and administer initial therapy.</p>