YerevaNN/mimic3-benchmarks
Python suite to construct benchmark machine learning datasets from the MIMIC-III 💊 clinical database.
This tool helps medical researchers and healthcare data scientists prepare clinical data from the MIMIC-III database for machine learning. It takes raw patient records and transforms them into structured datasets for common inpatient prediction tasks like forecasting mortality, detecting decompensation, predicting length of stay, and classifying patient phenotypes. This is designed for researchers who want to apply machine learning to critical care data without the heavy lift of data engineering.
874 stars. No commits in the last 6 months.
Use this if you need pre-processed, benchmark-ready clinical datasets from MIMIC-III to train and evaluate machine learning models for critical care prediction tasks.
Not ideal if you need to analyze raw MIMIC-III data directly or perform tasks outside of the four defined prediction benchmarks.
Stars
874
Forks
342
Language
Python
License
MIT
Category
Last pushed
Apr 16, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/YerevaNN/mimic3-benchmarks"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
rvandewater/YAIB
🧪Yet Another ICU Benchmark: a holistic framework for the standardization of clinical prediction...
justin13601/ACES
[ICLR 2025] ACES: Automatic Cohort Extraction System for Event-Streams
myADLM/ADLM-2025-Data-Challenge
Build an AI Tool for Extracting Information from Laboratory Documentation
cconsta1/SexEst
SexEst is an open-source Streamlit web application for predicting biological sex from skeletal...
Mubarraqqq/gnn-challenge
This project focuses on predicting maternal-fetal health outcomes, specifically early detection...