EARLY PREDICTION OF SEPSIS USING MACHINE LEARNING ALGORITHM: A BRIEF CLINICAL PERSPECTIVE

Abstract

 Sepsis is a worldwide cause of death owing to infection and associated immune system response. In situations of septic shock, mortality rates are highest in both developed and underdeveloped countries. Sepsis is a medical illness that requires immediate medical attention but can be avoided with advance warning. Sepsis affects an estimated 30 million people worldwide, with more than 6 million people dying each year. Among them More than 4.2 million new born and children are at risk of contracting the disease. To treat Sepsis, hospitals spend $24 billion (about 13% of all healthcare spending in the United States). The importance of early identification of sepsis in improving sepsis outcomes cannot be overstated. Each hour of delay in treatment increases the risk of death by 4 to 8%. As a result, developing a good model that may be able to tackle this problem becomes urgent and critical. We want to analyse, assess, and design an algorithm that will help us address some of the problems we've had when attempting the project.