Reducing bias and increasing utility by federated generative modeling of medical images using a centralized adversary
Open accessThe DOI matches its Crossref record; 34 of the 34 references it cites were found in a registry or an index.
Abstract
A major roadblock in machine learning for healthcare is the inability of data to be shared broadly, due to privacy concerns. Privacy preserving synthetic data generation is increasingly being seen as a solution to this problem. However, since healthcare data often has significant site-specific biases, it has motivated the use of federated learning when the goal is to utilize data from multiple sites for machine learning model training. Here, we introduce FELICIA (FEderated LearnIng with a CentralIzed Adversary), a generative mechanism enabling collaborative learning. It is a generalized extension of the (local) PrivGAN mechanism allowing to take into account the diversity (non-IID) nature of the federated sites. In particular, we show how a site with limited and biased data could benefit from other sites while keeping data from all the sources private. FELICIA works for a large family of Generative Adversarial Networks (GAN) architectures including vanilla and conditional GANs as demonstrated in this work. We show that by using the FELICIA mechanism, a site with a limited amount of images can generate high-quality synthetic images with improved utility, while none of the sites need to provide access to their real data. The sharing happens solely through a central discriminator with access limited to synthetic data. We demonstrate these benefits on several realistic healthcare scenarios using benchmark image datasets (MNIST, CIFAR-10) as well as on medical images for the task of skin lesion classification. We show that the utility of synthetic images generated by FELICIA surpasses that of the data available locally and we demonstrate that it can correct the reduced utility of a biased subgroup within a class.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“HD-ai models prefer to utilize both unstructured text information and medical imaging data for accurate and unbiased analysis [115,154].”
From Artificial intelligence bias in medical system designs: a systematic review (2023) · cites this paper as [115]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Learning Multiple Layers of Features from Tiny Images
Krizhevsky, A. (2024) · doi:10.57702/zp44cu3g · Cited by 23,395
GAN(Generative Adversarial Nets)
(2017) · Journal of Japan Society for Fuzzy Theory and Intelligent Informatics · doi:10.3156/jsoft.29.5_177_2 · Cited by 19,270
Unsupervised representation learning with deep convolutional generative adversarial networks
Radford, A., Metz, L., & Chintala, S. (2015) · arXiv · doi:10.48550/arxiv.1511.06434 · Cited by 12,235 · cited 1×
“In this paper, we apply ourmechanism to three separate GANs belonging to this family: i) the original GAN ([10]), ii) DCGAN ([21]), iii) conditional GAN ([18]).”
MethodConditional Generative Adversarial Nets
Mirza, M., & Osindero, S. (2014) · arXiv · doi:10.48550/arxiv.1411.1784 · Cited by 8,863
Deep Learning with Differential Privacy
Abadi, M., Chu, A., Goodfellow, I., McMahan, H. B., Mironov, I., Talwar, K., & Zhang, L. (2016) · Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/2976749.2978318 · Cited by 6,578
(2017) · International Conference on Machine Learning · Cited by 6,382
The Algorithmic Foundations of Differential Privacy
Dwork, C., & Roth, A. (2014) · Foundations and Trends® in Theoretical Computer Science · doi:10.1561/0400000042 · Cited by 5,557
The future of digital health with federated learning
Rieke, N., Hancox, J., Li, W., Milletarì, F., Roth, H. R., Albarqouni, S., Bakas, S., Galtier, M. N., Landman, B. A., Maier-Hein, K., et al. (2020) · npj Digital Medicine · doi:10.1038/s41746-020-00323-1 · Cited by 4,015 · cited 1×
“Another approach is to train a model in a federated learning setting such that the data never has to be shared ([7, 11, 23, 24]).”
BackgroundThe Algorithmic Foundations of Differential Privacy
Dwork, C., & Roth, A. (2013) · doi:10.1561/9781601988195 · Cited by 3,856
The Future of Digital Health with Federated Learning
Rieke, N., Hancox, J., Li, W., Milletari, F., Roth, H., Albarqouni, S., Bakas, S., Galtier, M. N., Landman, B., Maier-Hein, K., et al. (2020) · arXiv · doi:10.48550/arxiv.2003.08119 · Cited by 3,495
Generative Adversarial Networks
(2022) · The Science of Deep Learning · doi:10.1017/9781108891530.013 · Cited by 2,402
Deep learning-enabled medical computer vision
Esteva, A., Chou, K., Yeung, S., Naik, N., Madani, A., Mottaghi, A., Liu, Y., Topol, E., Dean, J., & Socher, R. (2021) · npj Digital Medicine · doi:10.1038/s41746-020-00376-2 · Cited by 1,406
PATE-GAN: Generating Synthetic Data with Differential Privacy Guarantees
Jordon, J., Yoon, J., & van der Schaar, M. (2018) · International Conference on Learning Representations · Cited by 1,086
Differentially Private Federated Learning: A Client Level Perspective
Geyer, R. C., Klein, T., & Nabi, M. (2017) · arXiv · doi:10.48550/arxiv.1712.07557 · Cited by 1,052
Learning Differentially Private Recurrent Language Models
McMahan, H. B., Ramage, D., Talwar, K., & Zhang, L. (2017) · arXiv · doi:10.48550/arxiv.1710.06963 · Cited by 849
Generative Adversarial Networks (GANs): Challenges, Solutions, and Future Directions
Saxena, D., & Cao, J. (2022) · ACM Computing Surveys · doi:10.1145/3446374 · Cited by 689
Semi-supervised Knowledge Transfer for Deep Learning from Private Training Data
Papernot, N., Abadi, M., Erlingsson, Ú., Goodfellow, I., & Talwar, K. (2016) · arXiv · doi:10.48550/arxiv.1610.05755 · Cited by 492
Privacy-Preserving Generative Deep Neural Networks Support Clinical Data Sharing
Beaulieu-Jones, B. K., Wu, Z. S., Williams, C., Lee, R., Bhavnani, S. P., Byrd, J. B., & Greene, C. S. (2019) · Circulation: Cardiovascular Quality and Outcomes · doi:10.1161/circoutcomes.118.005122 · Cited by 390
Differentially Private Generative Adversarial Network
Xie, L., Lin, K., Wang, S., Wang, F., & Zhou, J. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1802.06739 · Cited by 295
FedGAN: Federated Generative Adversarial Networks for Distributed DataPreprint
Rasouli, M., Sun, T., & Rajagopal, R. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2006.07228 · Cited by 227
MD-GAN: Multi-Discriminator Generative Adversarial Networks for Distributed Datasets
Hardy, C., Le Merrer, E., & Sericola, B. (2019) · 2019 IEEE International Parallel and Distributed Processing Symposium (IPDPS) · doi:10.1109/ipdps.2019.00095 · Cited by 201
Generalization and Equilibrium in Generative Adversarial Nets (GANs)
Arora, S., Ge, R., Liang, Y., Ma, T., & Zhang, Y. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1703.00573 · Cited by 182
Synthesis of diagnostic quality cancer pathology images by generative adversarial networks
Levine, A. B., Peng, J., Farnell, D., Nursey, M., Wang, Y., Naso, J. R., Ren, H., Farahani, H., Chen, C., Chiu, D., et al. (2020) · The Journal of Pathology · doi:10.1002/path.5509 · Cited by 120
Detect and correct bias in multi-site neuroimaging datasets
Wachinger, C., Rieckmann, A., & Pölsterl, S. (2021) · Medical Image Analysis · doi:10.1016/j.media.2020.101879 · Cited by 115
GS-WGAN: A Gradient-Sanitized Approach for Learning Differentially Private Generators
Chen, D., Orekondy, T., & Fritz, M. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2006.08265 · Cited by 73
Federated Generative Adversarial Learning
Fan, C., & Liu, P. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-60636-7_1 · Cited by 65
Private selection from private candidates
Liu, J., & Talwar, K. (2019) · Proceedings of the 51st Annual ACM SIGACT Symposium on Theory of Computing · doi:10.1145/3313276.3316377 · Cited by 64
Triastcyn, A., & Faltings, B. (2020) · IEEE Intelligent Systems · doi:10.1109/mis.2020.2993966 · Cited by 62
privGAN: Protecting GANs from membership inference attacks at low cost to utility
Mukherjee, S., Xu, Y., Trivedi, A., Patowary, N., & Ferres, J. L. (2021) · Proceedings on Privacy Enhancing Technologies · doi:10.2478/popets-2021-0041 · Cited by 38
Self-citationDifferential Privacy for Classifier Evaluation
Boyd, K., Lantz, E., & Page, D. (2015) · ACM Workshop on Artificial Intelligence and Security (AISec) · doi:10.1145/2808769.2808775 · Cited by 30
LOGAN: Membership Inference Attacks Against Generative ModelsPreprint
Hayes, J., Melis, L., Danezis, G., & Cristofaro, E. D. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1705.07663 · Cited by 20
MACE: A Flexible Framework for Membership Privacy Estimation in Generative Models
Xu, Y., Mukherjee, S., Liu, X., Tople, S., Dodhia, R., & Ferres, J. L. (2022) · Transactions on machine learning research · doi:10.48550/arxiv.2009.05683 · Cited by 11
Self-citationprivGAN: Protecting GANs from membership inference attacks at low costPreprint
Mukherjee, S., Xu, Y., Trivedi, A., & Ferres, J. L. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.2001.00071 · Cited by 10
Self-citationPrivate data sharing between decentralized users through the privGAN architecture
Rajotte, J.-F., & Ng, R. T. (2020) · IEEE International Enterprise Distributed Object Computing Conference Workshops (EDOCW) · doi:10.1109/edocw49879.2020.00018 · Cited by 4
Self-citation
Research works printed but not linked: 1
- Philipp Tschandl, Cliff Rosendahl, and Harald Kittler. "The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions". In: Scientific data 5.1 (2018), pp. 1-9.
Cited by
Algorithmic fairness in artificial intelligence for medicine and healthcare
Chen, R. J., Wang, J. J., Williamson, D. F. K., Chen, T. Y., Lipkova, J., Lu, M. Y., Sahai, S., & Mahmood, F. (2023) · Nature Biomedical Engineering · doi:10.1038/s41551-023-01056-8 · Cited by 750
Recent methodological advances in federated learning for healthcare
Zhang, F., Kreuter, D., Chen, Y. K., Dittmer, S., Tull, S., Shadbahr, T., Schut, M. C., Asselbergs, F. W., Kar, S., Sivapalaratnam, S., et al. (2024) · Patterns · doi:10.1016/j.patter.2024.101006 · Cited by 188
A survey of recent methods for addressing AI fairness and bias in biomedicine
Yang, Y., Lin, M., Zhao, H., Peng, Y., Huang, F., & Lu, Z. (2024) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2024.104646 · Cited by 146
Synthetic data as an enabler for machine learning applications in medicine
Rajotte, J.-F., Bergen, R., Buckeridge, D. L., El Emam, K., Ng, R., & Strome, E. (2022) · iScience · doi:10.1016/j.isci.2022.105331 · Cited by 124
Self-citationAdversarial Attacks Against Deep Generative Models on Data: A Survey
Sun, H., Zhu, T., Zhang, Z., Jin, D., Xiong, P., & Zhou, W. (2023) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2021.3130903 · Cited by 90
“[121] also built a federated learning structure with a central discriminator and multiple local generators and discriminators, motivated by the privGAN architecture [80].”
Background
Metadata
- Authors
- Rajotte, J.-F., Mukherjee, S., Robinson, C., Ortiz, A., West, C., Ferres, J. M. L., & Ng, R. T.
- Affiliations
- Data Science Institute, University of British Columbia, Vancouver, British Columbia, Canada
- Microsoft (United States) (US)
- University of British Columbia (CA)
- AI for Good Research Lab, Microsoft, Redmond, USA
- Venue
- Proceedings of the Conference on Information Technology for Social Good
- Publisher
- ACM
- Published
- 2021-09-09
- Pages
- 79-84
- OpenAlex
- W3125360523
- Cited by
- 34
- References
- 34
- Citations by year
- 2026: 1 · 2025: 5 · 2024: 10 · 2023: 9 · 2022: 5 · 2021: 4
- FWCI
- 3.23
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- International Conference on Information Technology for Social Good (GoodIT) (publishedVersion)
- arXiv (Cornell University) (submittedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University) (cc-by)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Generative Adversarial Networks and Image Synthesis
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Keywords
- federated learning, generative adversarial networks, synthetic data generation, privacy-preserving data sharing, non-IID data, medical image synthesis, skin lesion classification, Felicia, conditional generative adversarial networks
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.