Fairboard: a quantitative framework for equity assessment of healthcare models

TL;DR

This paper introduces Fairboard, a framework and dashboard for assessing and monitoring equity in AI medical models, revealing patient-specific biases and factors affecting performance.

Contribution

It provides a comprehensive evaluation of model equity across multiple dimensions and releases an open-source tool for ongoing fairness assessment in healthcare AI.

Findings

Patient identity explains more performance variance than model choice.

Clinical factors predict segmentation accuracy more strongly than architecture.

Newer models tend toward greater equity but lack formal fairness guarantees.

Abstract

Despite there now being more than 1,000 FDA-authorised AI medical devices, formal equity assessments -- whether model performance is uniform across patient subgroups -- are rare. Here, we evaluate the equity of 18 open-source brain tumour segmentation models across 648 glioma patients from two independent datasets (n = 11,664 model inferences) along distinct univariate, Bayesian multivariate, spatial, and representational dimensions. We find that patient identity consistently explains more performance variance than model choice, with clinical factors, including molecular diagnosis, tumour grade, and extent of resection, predicting segmentation accuracy more strongly than model architecture. A voxel-wise spatial meta-analysis identifies neuroanatomically localised biases that are compartment-specific yet often consistent across models. Within a high-dimensional latent space of lesion masks and clinic-demographic features, model performance clusters significantly, indicating that the patient feature space contains axes of algorithmic vulnerability. Although newer models tend toward greater equity, none provide a formal fairness guarantee. Lastly, we release Fairboard, an open-source, no-code dashboard that lowers barriers to equitable model monitoring in medical imaging.