Sepsis Prediction

Abstract

The intention of the proposed research is to create a machine learning model that predicts deadly disease at the earliest using  the real time  data and also to extend the research into a product-based approach. Generally, when an infection occurs in a  body,  when the  immune system helps to overcome the infection by producing antibodies, but sometimes  unusually  the immune system overreacts to the infection. This will  lead to a condition  called Sepsis. India ranks in south asia in Sepsis death rate. The death rate of sepsis is 213 per  lakh  people  in  India.  The major cause of death in the USA are Severe sepsis and septic shock. The early prediction of sepsis helps to quick recovery of the person. On the other hand, sepsis is very difficult to diagnose. The proposed solution contains the different methods of feature analysis and comparison between the classification model used. It can also reduce the death rate upto 13 per lakh people in India.