A Novel Method for Dengue Management Based on Vital Signs and Blood Profile
S.H.U.Briyatis1, S. C. Premaratne2, D.G.Harendra De Silva3
1S. H. U. Briyatis, Department of Information Technology, University of Moratuwa.
2S. C. Premaratne, Department of Information Technology, University of Moratuwa.
3D. G. Harendra De Silva, Department of Paediatrics, Faculty of Medicine, University of Colombo.
Manuscript received on 27 September 2019 | Revised Manuscript received on 09 November 2019 | Manuscript Published on 22 November 2019 | PP: 154-159 | Volume-8 Issue-6S3 September 2019 | Retrieval Number: F10250986S319/19©BEIESP | DOI: 10.35940/ijeat.F1025.0986S319
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Dengue a mosquito-borne viral disease has become a significant cause of morbidity and mortality in Sri Lanka over the past few years. Based on the symptoms signs and investigations, dengue can be classified as Dengue Fever (DF), Dengue Hemorrhagic Fever (DHF) and Dengue Shock Syndrome (DSS). Since it is a viral infection, medications have not been developed yet. Attempts to develop a vaccine in the prevention has not been completed yet. Our research is based on Cased-Based Reasoning (CBR) to develop an Online decision support system to manage dengue illness. CBR is one method that is capable of reasoning or solving problems on a case that has existed as a solution to a new problem. The CBR system identified the most critical vital signs, parameters, and investigations in different possible situations in DHF and DSS and designed six distinct cases. Clinical parameters and values or ranges of those parameters are used for this research with the guidance of Physicians. The system is developed based on index cases and rules. The system will predict the current situation of the patient by analyzing his/her past and present vital signs and investigations. Usually, DHF patients are monitored hourly, and important monitoring parameters will be entered into the system. The system will then display the current clinical stage and impending issues. Suggestions in the management are indicated. Using this system medical personnel can see how is the Pulse Pressure, Urine Output, Packed Cell Volume (PCV), Platelet Count and White Cell Count (WCC) vary according to the clinical stage of a particular patient. The evaluation of the system was carried out using some past cases of dengue in different stages. Our result shows that the CBR approach can gain significant accuracy in dengue management.
Keywords: Decision Support System, Case-Based Reasoning (CBR), Machine Learning, Prediction.
Scope of the Article: Probabilistic Models and Methods