DZC DIAG: mobile application based on expert system to aid in the diagnosis of dengue, Zika, and chikungunya
- PDF / 2,278,701 Bytes
- 16 Pages / 595.276 x 790.866 pts Page_size
- 101 Downloads / 164 Views
ORIGINAL ARTICLE
DZC DIAG: mobile application based on expert system to aid in the diagnosis of dengue, Zika, and chikungunya Adriana Peter Rodrigues de Araújo 1 & Maria Carolina Macedo de Araujo 2 & Thiago Coutinho Cavalcanti 3 & Cláudia Fernanda de Lacerda Vidal 4 & Marilú Gomes Netto Monte da Silva 2 Received: 22 November 2019 / Accepted: 18 July 2020 # International Federation for Medical and Biological Engineering 2020
Abstract Dengue, Zika, and chikungunya are epidemic diseases transmitted by the Aedes mosquito. These virus infections can be so severe to the point of bringing on mobility and neurological problems, or even death. Expert systems (ES) can be used as tools for the identification of patterns intended to solve problems in the same way as a professional specialist would. This work aimed to develop an ES in the form of an Android application to serve as a supportive tool in the diagnosis of these arboviruses. The goal is to associate the set of symptoms from a patient to a score related to the likelihood of them having these diseases. To make this possible, we implemented a rule-based ES which considers the presence of symptoms itself and the relation between them to associate the case under analysis to others found in the literature. We performed 96 tests (32 for each illness), and our system had a success rate of 96.88%. Resident physicians of a public hospital also analyzed these clinical cases and achieved an average success rate of 72.92%. Comparing the results of the method proposed and errors made by health professionals, we showed an improvement in the effectiveness of clinical diagnoses. Keywords Chikungunya virus . Dengue . Expert system . mHealth . Zika virus
1 Introduction Expert systems (ES) are computational systems based on artificial intelligence (AI). They are developed with the goal of solving problems similarly as a specialist would [1]. Expert systems are necessary because of various technological and economic-social factors, among which the following stand out [2]: difficulty of access to human experts in specific regions; availability for storage and formalization of knowledge
* Marilú Gomes Netto Monte da Silva [email protected] 1
Biomedical Engineering Department, University of Southern California, Los Angeles, USA
2
Biomedical Engineering Department, Federal University of Pernambuco, Av. da Arquitetura, s/n – Cidade Universitária, Recife, PE CEP: 50740-550, Brazil
3
Information and Communication Engineering Department, Daegu Gyeongbuk Institute of Science and Technology, Daegu, South Korea
4
Tropical Medicine Department, Federal University of Pernambuco, Recife, Brazil
deriving from multiple human experts; possibility to use a tool to support experts to make a decision or for professional training; and high cost and long time for professionals to specialize. In the 1970s, the first emerging expert systems that achieved their goals were DENDRAL [3], MYCIN [4], and PROSPECTOR [5]. The medical area has been one of the most benefited by expert systems because it
Data Loading...
