ARTIFICIAL INTELLIGENCE MODEL AS PREDICTOR FOR DENGUE OUTBREAKS

Dhesi Baha Raja; Rainier Mallol; Choo Yee Ting; Fadzilah Kamaludin; Rohani Ahmad; Suzilah Ismail; Vivek Jason Jayaraj; Bala Murali Sundram

doi:10.37268/mjphm/vol.19/no.2/art.176

Dhesi Baha Raja Public Health Medicine Specialist, Ministry of Health, 62590 Putrajaya, Malaysia
Rainier Mallol Computer Engineer, Pontifical Catholic University Mother and Master (Pontificia Universidad Católica Madre y Maestra), 51000 Santiago de los Caballeros, Dominican Republic
Choo Yee Ting Associate Professor, Faculty of Computing and Informatics, Multimedia University (MMU), 63100 Cyberjaya, Malaysia
Fadzilah Kamaludin Director, Institute for Medical Research (IMR), Ministry of Health, 50588 Kuala Lumpur, Malaysia
Rohani Ahmad Medical Entomology Unit, Infectious Diseases Research Centre, Institute for Medical Research (IMR), Ministry of Health, 50588 Kuala Lumpur, Malaysia
Suzilah Ismail SQS Statistical Consulting, School of Quantitative Sciences(SQS), Universiti Utara Malaysia (UUM), 06010 Sintok, Kedah, Malaysia
Vivek Jason Jayaraj Department of Social and Preventive Medicine, University Malaya, 50603 Kuala Lumpur, Malaysia
Bala Murali Sundram Public Health Medicine Specialist, Environmental Health Research Centre, Institute for Medical Research (IMR), Ministry of Health, 50588 Kuala Lumpur, Malaysia

DOI: https://doi.org/10.37268/mjphm/vol.19/no.2/art.176

Keywords: Aedes, Aegypti, Albopictus, C#, Bayesian Network, Dengue, Predictive Model

Abstract

Dengue is an increasing threat in Malaysia, particularly in the more densely populated regions of the country. We present an Artificial Intelligence driven model in predicting Aedes outbreak, using predictors of weather variables and vector indices sourced from the Ministry of Health. Analysis and predictions to estimate Aedes populations were conducted, with its results being used to infer the possibility of dengue outbreaks at pre-determined localities around the Klang Valley, Malaysia. A Bayesian Network machine learning technique was employed, with the model being trained using predictor variables such as temperature, rainfall, date of onset and notification, and vector indices such as the Ae. albopictus count, Ae. aegypti count and larval count. The interfaces of the system were developed using the C# language for Server-side configuration and programming, and HTML, CSS and JavaScript for the Client Side programming. The model was then used to predict the population of Aedes at periods of 7, 14, and 30 days. Using the Bayesian Network technique utilising the above predictor variables we proposed a finalised model with predictive accuracy ranging from 79%-84%. This model was developed into a Graphical User Interface, which was purposed to assist and educate the general public of regions at risk of developing dengue outbreak. This remains a valuable case-study on the importance of public data in the context of combating a public health risk via the development of models for predicting outbreaks of dengue which will hopefully spur further sharing of data by all parties in combating public health threats.

References

1. El Adlouni S, Beaulieu C, Ouarda TB, Gosselin PL, Saint-Hilaire A. Vector Borne Disease. Int J Health Geogr. 2007;6(1):40.

2. Brady OJ, Smith DL, Scott TW, Hay SI. Dengue disease outbreak definitions are implicitly variable. Epidemics. 2015 Jun;11:92–102.

3. Packierisamy PR, Ng C-W, Dahlui M, Inbaraj J, Balan VK, Halasa YA, et al. Cost of Dengue Vector Control Activities in Malaysia. Am J Trop Med Hyg. 2015 Nov;93(5):1020–7.

4. Chaudhry B, Wang J, Wu S, Maglione M, Mojica W, Roth E, et al. Systematic Review: Impact of Health Information Technology on Quality, Efficiency, and Costs of Medical Care. Ann Intern Med. 2006 May 16;144(10):742.

5. Drury P. The eHealth agenda for developing countries. World Hosp Health Serv. 2005;41(4):38–40.

6. Delone WH, Mclean ER. Information Systems Success: The Quest for the Dependent Variable. MIS Q. 1988;12:259–74.

7. Doll WJ, Torkzadeh G. The Measurement of End-User Computing Satisfaction: Theoretical and Methodological Issues. MIS Q. 1991 Mar;15(1):5.

8. Davis FD. Perceived Usefulness, Perceived Ease of Use, and User Acceptance of Information Technology. MIS Q. 1989 Sep;13(3):319.

9. Zaichkowsky JL. Measuring the Involvement Construct. J Consum Res. 1985 Dec 1;12(3):341.

10. Myers MD. Qualitative Research in Information Systems. MIS Q. 1997 Jun;21(2):241.

11. Gable GG, Sedera D, Chan T. Enterprise Systems Success: A Measurement Model. Proc Twenty-Fourth Int Conf Inf Syst. 2003;576:145–76.

12. Sedera D, Gable G. A factor and structural equation analysis of the enterprise systems success measurement model. Int Conf Inf Syst. 2004;449–64.

13. Seddon PB, Graeser V, Willcocks LP. Measuring organizational Information System effectiveness: An Overview and Update of senior management perspectives. ACM SIGMIS Database. 2002 Jun 1;33(2):11–28.

14. Hair JF, Black WC, Babin BJ, Anderson RE. Multivariate data analysis - Pearson. Pearson Education Inc; 2010.

15. Seddon P, Kiew M-Y. A Partial Test and Development of Delone and Mclean’s Model of IS Success. Australas J Inf Syst. 1996 Nov 1;4(1).

16. Venkatesh V, Morris MG, Davis GB, Davis FD. User Acceptance of Information Technology: Toward a Unified View. Source MIS Q. 2003;27(3):425–78.

17. Wyatt J. Clinical data systems : Development and evaluation. Lancet. 1994 Dec 17;344(8938):1682–8.

18. Rockart JF (John F, DeLong DW. Executive support systems : the emergence of top management computer use. Dow Jones-Irwin; 1988. 280 p.

19. Koivunen M, Välimäki M, Koskinen A, Staggers N, Katajisto J. The impact of individual factors on healthcare staff’s computer use in psychiatric hospitals. J Clin Nurs. 2009 Apr 1;18(8):1141–50.