Inspiration
What it does
We chose the MIMIC III- Clinical Dataset to analyze clinical data and make predictions about patients with Sepsis. Sepsis can be defined as a “potentially life-threatening condition that can occur as a result of the body’s response to an infection.” We were interested in this dataset as we wanted to use exploratory data analysis to assist projects in the healthcare industry. Insights from analysis could help medical professionals and institutions in improving healthcare for patients with Sepsis.
How we built it
What we used in our project: SQL - to access the data from UCI ODIT Sepsis Analysis Python - Pandas library to create data frames and manipulate the data Matplotlib to visualize the data R - to visualize the data
What we learned
Data from medical reports indicates that approximately 24% of people who visited hospitals developed some case of sepsis. The major causes of Sepsis as indicated by analysis of data are Streptococcus group D, Sepsis Staphylococcus aureus, Escherichia coli, Gram negative bacteria, and Streptococcus bacteria. Among all the races, Middle Eastern and American Indian possessed the greatest proportion of sepsis leading to death while most Americans and Caribbean islanders recovered from sepsis . Occurrence of Sepsis is higher in males than in females across the different races. However, females have a slightly higher risk of death due to sepsis.
Log in or sign up for Devpost to join the conversation.