Who Does Your Algorithm Fail? Investigating Age and Ethnic Bias in the MAMA-MIA Dataset

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Parikh, Aditya, Das, Sneha, Feragen, Aasa
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914126963408896
author Parikh, Aditya
Das, Sneha
Feragen, Aasa
author_facet Parikh, Aditya
Das, Sneha
Feragen, Aasa
contents Deep learning models aim to improve diagnostic workflows, but fairness evaluation remains underexplored beyond classification, e.g., in image segmentation. Unaddressed segmentation bias can lead to disparities in the quality of care for certain populations, potentially compounded across clinical decision points and amplified through iterative model development. Here, we audit the fairness of the automated segmentation labels provided in the breast cancer tumor segmentation dataset MAMA-MIA. We evaluate automated segmentation quality across age, ethnicity, and data source. Our analysis reveals an intrinsic age-related bias against younger patients that continues to persist even after controlling for confounding factors, such as data source. We hypothesize that this bias may be linked to physiological factors, a known challenge for both radiologists and automated systems. Finally, we show how aggregating data from multiple data sources influences site-specific ethnic biases, underscoring the necessity of investigating data at a granular level.
format Preprint
id arxiv_https___arxiv_org_abs_2510_27421
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Who Does Your Algorithm Fail? Investigating Age and Ethnic Bias in the MAMA-MIA Dataset
Parikh, Aditya
Das, Sneha
Feragen, Aasa
Computer Vision and Pattern Recognition
Artificial Intelligence
Deep learning models aim to improve diagnostic workflows, but fairness evaluation remains underexplored beyond classification, e.g., in image segmentation. Unaddressed segmentation bias can lead to disparities in the quality of care for certain populations, potentially compounded across clinical decision points and amplified through iterative model development. Here, we audit the fairness of the automated segmentation labels provided in the breast cancer tumor segmentation dataset MAMA-MIA. We evaluate automated segmentation quality across age, ethnicity, and data source. Our analysis reveals an intrinsic age-related bias against younger patients that continues to persist even after controlling for confounding factors, such as data source. We hypothesize that this bias may be linked to physiological factors, a known challenge for both radiologists and automated systems. Finally, we show how aggregating data from multiple data sources influences site-specific ethnic biases, underscoring the necessity of investigating data at a granular level.
title Who Does Your Algorithm Fail? Investigating Age and Ethnic Bias in the MAMA-MIA Dataset
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2510.27421