Monitoring fairness in machine learning models that predict patient mortality in the ICU
This addresses fairness issues for ICU patients, but it is incremental as it applies existing fairness analysis methods to a specific domain.
The paper tackled the problem of fairness in machine learning models predicting patient mortality in the ICU by proposing a fairness monitoring approach, showing that it provides more detailed insights than traditional accuracy metrics alone.
This work proposes a fairness monitoring approach for machine learning models that predict patient mortality in the ICU. We investigate how well models perform for patient groups with different race, sex and medical diagnoses. We investigate Documentation bias in clinical measurement, showing how fairness analysis provides a more detailed and insightful comparison of model performance than traditional accuracy metrics alone.