Application of Machine Learning for the predictions of Epidemic Diseases
An Epidemic is the rapid spread of infectious diseases to a large number of people in a given population within a short period of time. For example, AIDS spread widely throughout the world in the 1980s, and since that time it has taken the lives of more than 25 million people. Ebola is also an example of an epidemic.
Collapse of population in Mexico during the 16th century. Cocoliztli epidemics usually occurred within two years of a major drought. The epidemic in 1576 occurred after a drought stretching from Venezuela to Canada.
The correlation between drought and the disease has been thought to be that population numbers of the Vesper mouse, a carrier of viral hemorrhagic fever, increased during the rains that followed the drought, as conditions improved. Between 1519 and 1600 the population of Mexico fell from 15–30 million down to about 2 million.
Dependence of Epidemics
Epidemics of infectious diseases generally depends on several factors which also depends upon the population of that area hence the epidemic may restrict to only one location due to these factors. If the epidemic spreads to other countries then it is termed as Pandemic. There are several changes that occur in the pathogens that may trigger the epidemic like Virulence and changes to the host susceptibility to the infectious agent. The climate of an area is also an important factor.
Use of Machine Learning for the predictions
Predicting these epidemics (Infectious Diseases) can save thousands of lives and can be very valuable to society. With the fast advancements of Big Data and Machine Learning, these problems can be solved. With accurate analysis of data can help in early disease detection and hence better health care can be given. With the huge computational power nowadays it would rather take less time than it should’ve taken 10 years ago.
The geographical factors, Climate, and Population distribution should be taken into consideration for the Machine Learning prediction Model. Given an area where an epidemic outbreak has occurred, the Machine Learning model would be able to predict the next epidemic outbreak prone area. This ML model would be beneficial to different health organizations so that they could have a track of these epidemic diseases and stop the widespread. The epidemic of infectious diseases depends on the ecology of the host.
The modeling can be done for a particular location and epidemic diseases because the conditions that governs the outbreak of the different epidemics is different and hence models can be trained according to the disease and area.
Data Sources
The Data Sources may include weather report of an area, the population density of an area, Economic Profiles, etc. The World Health Organization site can be used as the main data source. The Exploratory Data Analysis can be done on the data to get the insight of the data and the important features which have a significant correlation with epidemic diseases.(https://www.who.int/gho/en/)
