Skip to contents

This version of the mimic dataset has been cleaned by removing columns with more than 10% missing data, imputing remaining missing values with the median, and dropping columns highly correlated with the outcome. It is designed for use in fairness-aware machine learning tasks and streamlined analysis.

Usage

mimic_preprocessed

Format

A data frame with fewer variables than the original due to preprocessing. Number of rows: 1776.

See also