An explanation of Bayesian logic in diagnostics
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
.ipynb_checkpoints
.gitignore
ATA case study.ipynb
README.md
Vaccines in a differential dx.ipynb

README.md

Bayesian diagnostics

A test for a relatively frequent (1:100) condition has 97% specificity and 85% sensitivity. What is the likelihood that a person testing positive has the disease? Few would guess the answer is a little over 20%. Bayes' theorem is a powerful tool to understand causal reasoning, and in particular understand how physicians, epidemiologists, veterinarians and others in the healthcare field reason about the likelihood of a particular diagnosis.

This repo accompanies this blog post (oops... it hasn't been completed yet!) and is intended to illustrate the processes discussed therein.

Table of contents