As stated by Charles Darwin, the motive of all organisms is to evolve and adapt to the surrounding environment. It is always survaival of the fittest. For humans to survive, it is necessary we take necessary precautions even before a disaster strikes , be it natural or man made. One of the first things the government has to take care of after a natural disaster like flood strikes is to minimize the spread of epidemic diseasses. We propose to tackle the outbreaks of epidemics and it's adverse effects by predicting the possibility of an outbreak in the future and if there is a possibility, the extent to which it would affect life.
Epidemic refers to rapid spread of potentially infectious disease to a large number of people in a given population within a short period of time which adds to the complexity of the epidemics.
We plan to use machine learning algorithms to predict the outbreak of epidemic diseases and determine its propagation by taking into account various factors such as number of people infected, climatic conditions ,time taken for symptoms to show up, rate of spread of infection, time taken for the infection to subside, frequency of natural disasters and other factors. Unfortunately, just as the human body grows immune to various viruses and bacteria, these organisms also grow immune to the antibiotics used in the past and the climatic conditions. Hence, we need to consider the organism responsible for the outbreak too and it's past data. This would help the government to take necessary steps for prevention of outbreak by isolation of the infected masses or managing the after effects (if it's too late) by providing rapid medical aid to the potential victims to lower the death rate. This project aims at preventing further outbreak of epidemics by alerting people and government about the spread of diseases. We plan to predict the spread of specific communicable diseases for which sufficient and reliable data is available for many countries in the world, like cholera, malaria, zika and AIDS.
By observing data on factors like number of cases reported per year, number of deaths caused by the diseases per year, food deficit of previous years etc., we will be able to determine if a particular country is gaping at an epidemic.
We are planning to use climate datasets from WHO, and various disease outbreak data from reepidemics consortium, other disease datasets from Kaggle (eg: zika virus) and other sites as available.
- Global Health Observatory Data Repository: http://apps.who.int/gho/data/node.home
- Cause of death, by communicable diseases and maternal, prenatal and nutrition conditions (% of total): https://data.worldbank.org/indicator/SH.DTH.COMM.ZS
- Food Deficit: (https://data.worldbank.org/indicator/SN.ITK.DFCT?view=chart)
- Data for Zika: https://www.kaggle.com/cdc/zika-virus-epidemic
We will work with python and its various libraries available for machine learning along with other technologies like Microsoft Azure specified in the competition.
- Tejas R
- Tejas Kumar
- Darshan V
- Using climate to predict infectious disease epidemics: http://www.who.int/globalchange/publications/infectdiseases/en/
- www.who.int/globalchange/publications/infectdiseases.pdf?ua=1