PuSH - Publication Server of Helmholtz Zentrum München

Export: TXT Text RIS Endnote (RIS) BIB BibTeX
Meissen, F.* ; Breuer, S.* ; Knolle, M.* ; Buyx, A.* ; Müller, R.* ; Kaissis, G. ; Wiestler, B.* ; Rückert, D.*

(Predictable) performance bias in unsupervised anomaly detection.

EBioMedicine 101:105002 (2024)
Publ. Version/Full Text DOI PMC
Open Access Gold
Creative Commons Lizenzvertrag
BACKGROUND: With the ever-increasing amount of medical imaging data, the demand for algorithms to assist clinicians has amplified. Unsupervised anomaly detection (UAD) models promise to aid in the crucial first step of disease detection. While previous studies have thoroughly explored fairness in supervised models in healthcare, for UAD, this has so far been unexplored. METHODS: In this study, we evaluated how dataset composition regarding subgroups manifests in disparate performance of UAD models along multiple protected variables on three large-scale publicly available chest X-ray datasets. Our experiments were validated using two state-of-the-art UAD models for medical images. Finally, we introduced subgroup-AUROC (sAUROC), which aids in quantifying fairness in machine learning. FINDINGS: Our experiments revealed empirical "fairness laws" (similar to "scaling laws" for Transformers) for training-dataset composition: Linear relationships between anomaly detection performance within a subpopulation and its representation in the training data. Our study further revealed performance disparities, even in the case of balanced training data, and compound effects that exacerbate the drop in performance for subjects associated with multiple adversely affected groups. INTERPRETATION: Our study quantified the disparate performance of UAD models against certain demographic subgroups. Importantly, we showed that this unfairness cannot be mitigated by balanced representation alone. Instead, the representation of some subgroups seems harder to learn by UAD models than that of others. The empirical "fairness laws" discovered in our study make disparate performance in UAD models easier to estimate and aid in determining the most desirable dataset composition. FUNDING: European Research Council Deep4MI.
Impact Factor
Scopus SNIP
Altmetric
9.700
0.000
Tags
Annotations
Special Publikation
Hide on homepage

Edit extra information
Edit own tags
Private
Edit own annotation
Private
Hide on publication lists
on hompage
Mark as special
publikation
Publication type Article: Journal article
Document type Scientific Article
Keywords Algorithmic Bias ; Anomaly Detection ; Artificial Intelligence ; Machine Learning ; Subgroup Disparities; Bare Bones
Language english
Publication Year 2024
HGF-reported in Year 2024
ISSN (print) / ISBN 2352-3964
e-ISSN 2352-3964
Journal EBioMedicine
Quellenangaben Volume: 101, Issue: , Pages: , Article Number: 105002 Supplement: ,
Publisher Elsevier
Publishing Place Amsterdam [u.a.]
Reviewing status Peer reviewed
Institute(s) Institute for Machine Learning in Biomed Imaging (IML)
POF-Topic(s) 30205 - Bioengineering and Digital Health
Research field(s) Enabling and Novel Technologies
PSP Element(s) G-507100-001
Grants European Research Council
DOI 10.1016/j.ebiom.2024.105002
WOS ID 001184414100001
PubMed ID 38335791
Erfassungsdatum 2024-04-19
[➜Report correction]