文章基本信息

标题：DATA-DRIVEN NEURAL NETWORK MODEL FOR EARLY SELF-DIAGNOSIS OF DENGUE SYMPTOMS
本地全文：下载
作者：NOR AZURA HUSIN ; NURUL IMAN SAIFUL BAHARI ; HAZLINA HAMDAN 等
期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2020
卷号：98
期号：24
页码：4228-4238
出版社：Journal of Theoretical and Applied
摘要：Dengue fever is one of the main public health concerns and endemic diseases in many countries, especially in tropical and subtropical regions. In severe cases, people infected with dengue may experience severe bleeding which may lead to death if the infection is not properly treated. It is a standard procedure when a person is referred to the hospital with a high fever for more than two days to be required to undergo a dengue fever screening at the triage before further clinical tests are done to confirm the patient�s medical condition. Thus, an individual need to conduct an early self-diagnosis to identify the probability that he/she have been infected with dengue fever and further seek professional help from medical practitioners. There are many dengue fever symptoms outlined by the World Health Organization (WHO) such as sudden high fever, headache, abdominal pain, persistent vomiting, rapid breathing, bleeding gums, fatigue, restlessness, and vomiting blood, but the identification of the highly significant symptoms among the less significant symptoms are still scarce. Identification of significant symptoms from the dengue dataset may help patients and medical practitioners to acknowledge the alarming symptoms and endeavour for immediate action to prevent dengue outbreak and fatality. Hence, the identification of significant dengue symptoms may assist in system development to determine the weightage of each attribute in the system. This may result in better prediction of dengue. Therefore, the objective of this study is to develop an early self-diagnosis system using an artificial neural network with the ability to produce a reliable result based on the identification of significant symptoms. The model accuracy of 100% indicates the high reliability of the developed early self-diagnosis system. A mobile application was developed based on the prediction model for patients and medical practitioners as the target users. This study contributes to the field of public health by providing early detection for people who are at risk of being infected by dengue disease. The advancement in early detection technology brings such a huge positive impact in healthcare. Early identification of significant symptoms ensures that better focus can be given to the identified symptoms for a more reliable dengue fever assessment.
关键词：Dengue Fever;Diagnose;Early Self-Diagnosis;Machine Learning;Significant Symptoms.