Implementing Fairness in Real-World Healthcare Machine Learning through Datasheet for Database
Bibliographic record
Abstract
Healthcare Machine Learning (HML) models are revolutionizing the healthcare industry, \npromising improved patient outcomes and enhanced public health. However, it is \nessential to ensure fairness, i.e., models delivering equitable performance to all individuals, \nirrespective of their inherent or acquired characteristics. This requires a thorough \nexamination of the data used and the specific applications of these models. \nThis study conducted a six-year systematic survey of models trained on the Medical \nInformation Mart for Intensive Care (MIMIC) clinical research database (CRD) – one of \nthe most popular and widely used HML databases to explore the link between data and \nfairness in HML. \nThe results were striking: for the popular MIMIC IV – ICU mortality task, a naive baseline \noutperformed the state-of-the-art (SOTA) model in prediction performance, demonstrating \ngreater fairness across subgroups (while still somewhat unfair). These findings \ndemonstrate the urgent need to integrate fairness into healthcare machine learning models \nand a greater need to include practitioners in HML modeling. \nTo achieve this, we propose a data-centric approach to fairness through our ‘Datasheet \nfor MIMIC IV v2.0 CRD’, modeled after the recent works recommending datasheets for \ndatasets. Given that MIMIC is large and complex, this datasheet will assist practitioners in \nidentifying data anomalies and task-specific feature-target relationships during modeling, \nthereby fostering the development of equitable HML models.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.118 | 0.256 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.004 |
| Science and technology studies | 0.002 | 0.003 |
| Scholarly communication | 0.010 | 0.015 |
| Open science | 0.006 | 0.010 |
| Research integrity | 0.002 | 0.004 |
| Insufficient payload (model declined to judge) | 0.004 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".