Skip to content
Race and Mistrust in End-of-Life Care
Jupyter Notebook
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.


Race and Mistrust in End-of-Life Care

Short workshop paper (FAT/ML 2018):

 title={Modeling Mistrust in End-of-Life Care},
 author={W. Boag and H. Suresh and L.A. Celi and P. Szolovits and M. Ghassemi},
 publisher={Fairness, Accountability, and Transparency in Machine Learning (FAT/ML 2018) workshop},

15-page conference paper (MLHC 2018):

 title =      {Racial Disparities and Mistrust in End-of-Life Care},
 author =      {Boag, W. and Suresh, H. and Celi, L.A. and Szolovits, P. and Ghassemi, M.},
 booktitle =      {Proceedings of the 3rd Machine Learning for Healthcare Conference},
 pages =      {587--602},
 year =      {2018},
 volume =      {85},
 series =      {Proceedings of Machine Learning Research},
 address =      {Palo Alto, California},
 month =      {17--18 Aug},
 publisher =      {PMLR},
 pdf =      {},
 url =      {},

Masters Thesis:

 title={Quantifying Racial Disparities in End-of-Life Care},
 author={W. Boag},

The code folder has 6 notebooks:

1. race_mimic_aggressive.ipynb: Generate the figures for race-based treatment disparities in MIMIC

2. trust.ipynb: Generates the various mistrust metric proxies and saves them to file.

3. mistrust_mimic_aggressive.ipynb: Generate the figures for mistrust-based treatment disparities in MIMIC

4. cohort.ipynb: Generate additional stats (e.g. table one, pairwise comparisons of metrics & severity score, etc)

5. outcomes_ml.ipynb: Uses trust-based features to improve predictions for clinical tasks.

6. race_eicu_aggressive.ipynb: Generate the figures for race-based treatment disparities in eICU

Run trust.ipynb before running:

- mistrust_mimic_aggressive.ipynb

- cohort.ipynb

- outcomes_ml.ipynb
You can’t perform that action at this time.