A systematic study of race and sex bias in CNN-based cardiac MR segmentation

TL;DR

This study systematically investigates how training data imbalance affects race and sex bias in CNN-based cardiac MRI segmentation, revealing significant racial bias but negligible sex bias, emphasizing the importance of balanced datasets.

Contribution

It is the first comprehensive analysis of demographic bias in CNN cardiac MRI segmentation, demonstrating the impact of data imbalance on racial bias.

Findings

Significant racial bias observed in CNN segmentation models.

No significant sex bias detected in the models.

Highlights the importance of balanced demographic representation in medical imaging datasets.

Abstract

In computer vision there has been significant research interest in assessing potential demographic bias in deep learning models. One of the main causes of such bias is imbalance in the training data. In medical imaging, where the potential impact of bias is arguably much greater, there has been less interest. In medical imaging pipelines, segmentation of structures of interest plays an important role in estimating clinical biomarkers that are subsequently used to inform patient management. Convolutional neural networks (CNNs) are starting to be used to automate this process. We present the first systematic study of the impact of training set imbalance on race and sex bias in CNN-based segmentation. We focus on segmentation of the structures of the heart from short axis cine cardiac magnetic resonance images, and train multiple CNN segmentation models with different levels of race/sex imbalance. We find no significant bias in the sex experiment but significant bias in two separate race experiments, highlighting the need to consider adequate representation of different demographic groups in health datasets.