PuSH - Publication Server of Helmholtz Zentrum München

Export: TXT Text RIS Endnote (RIS) BIB BibTeX
Tayebi Arasteh, S.* ; Lotfinia, M.* ; Nolte, T.* ; Sähn, M.J.* ; Isfort, P.* ; Kühl, C.* ; Nebelung, S.* ; Kaissis, G. ; Truhn, D.*

Securing collaborative medical aI by using differential privacy: Domain transfer for classification of chest radiographs.

Radiol. Artif. Intell. 6, DOI: 10.1148/ryai.230212 (2024)
Publ. Version/Full Text DOI
Closed
Open Access Green as soon as Postprint is submitted to ZB.
Purpose: To investigate the integration of differential privacy (DP) and analyze its impact on model performance as compared with models trained without DP. Materials and Methods: Leveraging more than 590 000 chest radiographs from five institutions, including VinDr-CXR from Vietnam, ChestX-ray14 and CheXpert from the United States, UKA-CXR from Germany, and PadChest from Spain, the authors evaluated the efficacy of DP-enhanced domain transfer (DP-DT) in classifying cardiomegaly, pleural effusion, pneumonia, atelectasis, and healthy individuals. Diagnostic performance and sex-specific and age-specific demographic fairness of DP-DT and of non–DP-DT models were compared using the area under the receiver operating characteristic curve (AUC) as the main metric, as well as accuracy, sensitivity, and specificity as secondary metrics, and evaluated for statistical significance using paired Student t tests. Results: Even with high privacy levels (ε ≈ 1), DP-DT showed no evidence of differences compared with non–DP-DT in terms of a decrease in AUC of cross-institutional performance as compared with single-institutional performance (VinDr-CXR: 0.07 vs 0.07, P = .96; ChestX-ray14: 0.07 vs 0.06, P = .12; CheXpert: 0.07 vs 0.07, P = .18; UKA-CXR: 0.18 vs 0.18, P = .90; and PadChest: 0.07 vs 0.07, P = .35). Furthermore, AUC differences between DP-DT and non–DP-DT models were less than 1% for all sex subgroups (P > .33 for female and P > .22 for male, for all domains) and nearly all age subgroups (P > .16 for younger participants, P > .33 for adults, and P > .27 for older adults, for nearly all domains). Conclusion: Cross-institutional performance of artificial intelligence models was not affected by DP.
Impact Factor
Scopus SNIP
Web of Science
Times Cited
Altmetric
8.100
0.000
1
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 Chest Radiograph ; Computer Applications–general ; Computer-aided Diagnosis ; Convolutional Neural Network (cnn) ; Deep Learning ; Diagnosis ; Differential Privacy ; Domain Transfer ; Forensics ; Image Postprocessing ; Informatics ; Neural Networks ; Privacy-preserving Ai ; Supervised Learning ; Thorax ; Transfer Learning
Language english
Publication Year 2024
Prepublished in Year 2023
HGF-reported in Year 2023
ISSN (print) / ISBN 2638-6100
e-ISSN 2638-6100
Journal Radiology. Artificial intelligence
Quellenangaben Volume: 6, Issue: 1 Pages: , Article Number: , Supplement: ,
Publisher Radiological Society of North America
Publishing Place 820 Jorie Blvd, Suite 200, Oak Brook, Illinois, United States
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 Bavarian State Ministry for Science and the Arts
BMBF
Radiological Cooperative Network (RACOON) under the German Federal Ministry of Education and Research (BMBF)
DOI 10.1148/ryai.230212
WOS ID 001171836700002
Scopus ID 85184418732
Erfassungsdatum 2024-02-20
[➜Report correction]