Framework of Predicting the Acute Hepatitis C Outcomes By Using Data Mining Techniques | ||||
| ERJ. Engineering Research Journal | ||||
| Article 15, Volume 45, Issue 4, October 2022, Page 657-664 PDF (937.16 K) | ||||
| Document Type: Original Article | ||||
| DOI: 10.21608/erjm.2022.150105.1191 | ||||
 View on SCiNiTO | ||||
| Authors | ||||
Ahmed Abdel-Qader  1; Arabi Keshk2; Sanaa M. Kamal3
| ||||
| 1Faculty of Computers and Information, Menoufia University | ||||
| 2Computer Science, Faculty of Computers and Information, Menoufia University, Egypt | ||||
| 3Department of Gastroenterology and Hepatology, Faculty of Medicine, Ain Shams University, Cairo, Egypt | ||||
| Abstract | ||||
| Hepatitis C is a common disease that attacks the human liver. The hepatitis C infection could evolve into chronic hepatitis in almost 80% of cases. The acute stage of the C virus presents a turning point in the development of hepatitis C. Due to the lack of guidelines, physicians are not able to decide on whether to pursue clinical procedures or not. Furthermore, no one knows if it had healed off-hand, or it will need treatment. In this paper, a prediction model had been created to predict the acute hepatitis c outcomes based on data mining methods using clinical data. The dataset was collected from different centers in Egypt and Europe in the text format. The model depends on a framework that consists of six main phases. The phases are problem understanding, data realizing, preprocessing, modeling, appraisal (evaluation), and Visualization. Decision tree technique is the used data mining method that can produce a decision tree (prediction model). This study introduce a developed application based on a knowledge base. The knowledge base used the rules of prediction model as an input for the developed application. Then, the outcomes were predicted to be an output from the application. The experimental results showed that the hepatitis c virus core antigen is a reliable method for monitoring disease cases. The core antigen is a reliable monitoring tool for treatment decision making. Also, the averages of the four models that had been obtained are 92.3% of sensitivity, 88.91% of specificity and 90.12 % of accuracy. | ||||
| Keywords | ||||
| Acute Hepatitis C Virus; Acute Hepatitis C Outcomes; Prediction Model; Data Mining Techniques; Decision Tree | ||||
|
Statistics Article View: 115 PDF Download: 93 |
||||
