Auditing fairness in clinical AI systems using provenance-based simulation: a comparative and regulatory…

StudyFrontiers in artificial intelligence· 2026

Auditing fairness in clinical AI systems using provenance-based simulation: a comparative and regulatory perspective.

Fidelis Alu, Sunkanmi Oluwadare, Nnennaya Ngwanma Halliday, Uchenna Emmanuel Agwunobi

Read paper

AI Evaluation & GovernanceStudyQ1IF 7.8

Random forest models showed 57% less gender bias than logistic regression despite lower accuracy

Auditing fairness in clinical AI systems using provenance-based simulation: a comparative and regulatory perspective.

Fidelis Alu … Uchenna Emmanuel Agwunobi

Frontiers in artificial intelligence · 2026 · Q1 in Artificial Intelligence

Background

The adoption of Artificial Intelligence (Artificial Intelligence) in clinical decision support has encountered obstacles due to algorithmic bias and lack of transparency.

Methods

We simulated audits on a synthetic patient dataset ( N = 1,000), comparing logistic regression and random forest models to detect gender bias using 5-fold crossvalidation and permutation testing.

n = 1,000

Results

random forest was much fairer across gender than logistic regression

Logistic regression

0.256EOD

Random forest

0.055EOD

More results

Logistic regression achieved 75.2 ± 1.0% accuracy (AUC = 0.806 ± 0.030) and random forest achieved 70.1 ± 1.4% accuracy (AUC = 0.745 ± 0.020).

Provenance logs successfully detected gender biases in both models.

More results

Logistic regression exhibited statistically significant bias (EOD = +0.256, p = 0.0080), while random forest's smaller disparity (EOD = +0.055, p = 0.5664) was not statistically significant, demonstrating that our audit distinguishes systematic discrimination from random variation.

“

Conclusion

This framework maps to FDA transparency guidelines and ONC HTI-1 requirements, demonstrating how provenance-based auditing supports regulatory compliance and provides a pathway toward more responsible and equitable Artificial Intelligence in clinical settings.

Auditing fairness in clinical AI systems using provenance-based simulation: a comparative and regulatory perspective.

Auditing fairness in clinical AI systems using provenance-based simulation: a comparative and regulatory perspective.

Related papers