Inspiration
We thought sepsis was very interesting, especially because of its relation to medicine through data science.
What it does
The project talks about features of the data that we looked into as well as some factors that could influence the occurrence of sepsis.
How we built it
We used SQL queries to clean and format the data, and created visuals using python and matlab.
Challenges we ran into
The database did not include a lot of information that we were looking for, and paired with the difficulty of tailoring and shaping the SQL datasets to be able to work with python, the creation of a plausible model to reasonably predict any sepsis was hard.
What we learned
Not all data can be conclusive, and this dataset was inconclusive, given its extremely biased nature with respect to racial inclusivity.
What's next for Does Race and Ethnicity Correlate with a Sepsis Diagnosis?
Sadly, there are not future steps, as the dataset was modified to be a sample set mimicking a real omop EHR.
Log in or sign up for Devpost to join the conversation.