Abstract:
Dengue fever is the most rapidly spreading mosquito-borne viral disease in the world. An estimated 50 million dengue infections occur annually and approximately 2.5 billion people live in dengue endemic countries, one of which is the Philippines. Dengue is often characterized with influenza-like symptoms which include fever, rash, and body pain. Carried by the Aedes Egypti mosquito, the disease spreads as mosquitoes feed on an infected host and is passed on and transmitted by feeding on another host. Reported cases significantly increase during the rainy season wherein stagnant water becomes home to the mosquitoes and their offspring.
Often, failure to properly diagnose the disease leads to further infection and in worst cases, fatality. Ergo, early detection and proper diagnosis can be critical and life-saving. Clinical decision support systems (CDSS) help achieve and attain that. CDSSs combine knowledge and data to generate and present helpful information to health care providers as care is being delivered.
The Genetic Neural Network Analytic Tool for Dengue Fever (GNAT) is devised and created to be a CDSS that provides a novel, efficient, and robust way to detect and diagnose dengue fever. GNAT utilizes artificial neural networks (ANN) and genetic algorithm (GA) to classify whether patients are infected with dengue fever. Inspired by the biological brain, ANNs emulate the signal integration and threshold firing of biological neurons with the use of mathematical models and statistical weights. Hampered by the lack of an efficient training method, GA is used to optimize the weights between artificial neurons. GAs, as it mimics evolutionary processes, provides the best if not the correct solution to a wide range of problems.
