The paper introduces FRAME, a two‑step framework for auditing fairness claims in medical imaging. First, it derives a fair‑model reference distribution that captures the portion of subgroup performance differences attributable to sampling variation. Second, it tests the remaining difference using operators in representation space to assess whether demographic information or disease entanglement drives the bias. Across a large dataset of 702,206 images and 36 encoders, the reference explains a substantial median share of race and age differences, while interventions such as injecting demographic decodability or entangling disease direction have limited impact on the residual bias.
By Mahshad Lotfinia, Daniel Truhn, Andreas Maier, Soroosh Tayebi Arasteh
The paper examines how fairness conclusions in ICU mortality prediction using MIMIC-IV depend on the choice of metrics and the granularity of demographic analysis. It compares predictive-utility and subgroup-error metrics across various fairness interventions and introduces a lightweight adaptation strategy that balances ethnicity, gender, and insurance representation without conditioning on mortality outcomes. The study finds that different interventions can be evaluated differently across accuracy, sensitivity, and false-positive rate, and that marginal demographic summaries may hide heterogeneous error patterns within intersectional subgroups.
By Abdullah Al Noman, Fahmid Al Rifat, Tahrima Hashem, Syed Muhammad Ibne Zulfiker, Rishov Paul, Tanzima HAshem
The paper introduces the Explanation Consistency Score (ECS), a fairness‑aware metric that uses Jensen‑Shannon divergence to measure how similar attribution maps are across demographic subgroups. Applied to diabetic retinopathy screening, ECS is evaluated both overall and within disease severity levels. Results show that although predictive performance varies among ethnic groups, explanation consistency remains high and is not significantly linked to performance disparities, indicating that predictive fairness and explanation consistency assess different aspects of model behavior.
By Kerol Djoumessi, Philipp Berens
arXiv:2607. 14984v1 Announce Type: new Abstract: Per-subgroup fairness audits of medical image classifiers face a sample-size problem: minority subgroups in held-out test sets have so few samples that the resulting confidence intervals on per-subgroup performance are wider than the bias the audit is meant to detect.
By Mahmoud Ibrahim, Bart Elen, Chang Sun, Gokhan Ertaylan, Michel Dumontier
arXiv:2607. 07852v1 Announce Type: cross Abstract: Automated segmentation of cervical-spine MRI is increasingly used in clinical workflows, yet no fairness audit exists for this anatomy.
By Linus Juni, Aasa Feragen, Aditya Parikh
The study examines how three parameter‑efficient adaptation methods—linear heads on the raw CLS token, an MLP, and an attention‑pooling module—affect pathology classification accuracy and subgroup fairness when applied to a frozen Rad‑DINO chest X‑ray encoder. Using the MIMIC‑CXR dataset, the authors evaluate eight pathologies across race, sex, and imaging‑view subgroups, finding that attention pooling yields the best overall performance and encodes protected attributes most strongly, yet higher performance does not consistently reduce subgroup disparities. The results show that attribute encoding strength and layer choice do not reliably predict fairness outcomes, indicating that fairness must be assessed directly for each task.
By Dhruv Gupta, Emma A. M. Stanley, Fabio De Sousa Ribeiro, Sujal R. Desai, Ben Glocker
arXiv:2607. 08953v1 Announce Type: new Abstract: Algorithmic fairness methods are increasingly used to identify and mitigate bias in machine learning models, yet most approaches are evaluated in isolation and along single demographic axes.
By Nick Souligne, Isabella Mixton-Garcia, Vignesh Subbian
arXiv:2604. 16450v2 Announce Type: replace-cross Abstract: Intersectional biases in healthcare data can produce compound disparities in clinical machine learning models, yet most fairness evaluations assess demographic attributes independently.
By Nick Souligne, Vignesh Subbian
FairLens is a benchmark and evaluation framework that measures fairness and validity of vision‑language models (VLMs) in high‑stakes domains such as hiring, legal, and healthcare. It uses over 100,000 face‑image and question pairs covering gender, race, and age, and assesses responses through demographic parity, soundness, demographic association, and bias in free‑text generation. The study finds that VLMs often make unwarranted inferences from faces rather than abstaining, especially in legal and healthcare contexts, and that small parity gaps can still hide unsafe treatment across groups.
By Vahid Reza Khazaie, Ahmed Y. Radwan, Shaina Raza
arXiv:2605.10521v2 Announce Type: replace-cross
Abstract: As medical AI expands across diverse healthcare settings worldwide, equitable performance across patient populations is becoming essential to...
By Yiqi Tian, Jinwoong Park, Sangjoon Park, Bo Zeng, Pengfei Jin, Yujin Oh, Quanzheng Li
FairGlucose is a 300‑patient CGM cohort balanced across 12 demographic strata, providing 132,480 forecasting samples and 3,945 behavioral events. Benchmarking 33 models on 2‑hour glucose forecasting revealed that population‑level validation masks significant subgroup disparities, with error ratios ranging from 0.8 to 1.4 and T1D patients experiencing 6 mg/dL higher error than T2D. The study shows that these gaps persist across all models, align with clinically hard cases, and vary with input‑length sensitivity, underscoring the need for subgroup‑disaggregated reporting in digital health AI.
By Junjie Luo, Xuzhe Zhi, Rui Han, Abhimanyu Kumbara, Anand K. Iyer, Mansur E. Shomali, Ritu Agarwal, Guodong Gordon Gao
Algorithmic fairness methods are increasingly used to identify and mitigate bias in machine learning models, yet most approaches are evaluated in isolation and along single demographic axes. This limits practical guidance for selecting fairness strategies, where disparities may arise across intersectional subgroups and across multiple stages of the modeling lifecycle.