In the Pursuit of Privacy: The Promises and Predicaments of Federated Learning in Healthcare
Open accessThe DOI matches its Crossref record; 22 of the 22 references it cites were found in a registry or an index.
Abstract
Artificial Intelligence and its subdomain, Machine Learning (ML), have shown the potential to make an unprecedented impact in healthcare. Federated Learning (FL) has been introduced to alleviate some of the limitations of ML, particularly the capability to train on larger datasets for improved performance, which is usually cumbersome for an inter-institutional collaboration due to existing patient protection laws and regulations. Moreover, FL may also play a crucial role in circumventing ML’s exigent bias problem by accessing underrepresented groups’ data spanning geographically distributed locations. In this paper, we have discussed three FL challenges, namely: privacy of the model exchange, ethical perspectives, and legal considerations. Lastly, we have proposed a model that could aide in assessing data contributions of a FL implementation. In light of the expediency and adaptability of using the Sørensen–Dice Coefficient over the more limited (e.g., horizontal FL) and computationally expensive Shapley Values, we sought to demonstrate a new paradigm that we hope, will become invaluable for sharing any profit and responsibilities that may accompany a FL endeavor.
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.
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
Dissecting racial bias in an algorithm used to manage the health of populations
Obermeyer, Z., Powers, B., Vogeli, C., & Mullainathan, S. (2019) · Science · doi:10.1126/science.aax2342 · Cited by 7,137 · cited 1×
“However, data availability and underrepresentation of minorities in healthcare datasets are well-known impediments to ML research (Obermeyer et al., 2019) and lead to relatively low performance for disproportionately represented ethnic groups (Gao and Cui, 2020).”
BackgroundFederated learning: Challenges, methods, and future directions
Li, T., Sahu, A. K., Talwalkar, A., & Smith, V. (2020) · IEEE Signal Processing Magazine · doi:10.1109/msp.2020.2975749 · Cited by 6,210 · cited 1×
“GDP is the most prevalent method of privacy protection due to its simple algorithmic convenience, information guarantee, and relatively small overhead cost (Li et al., 2020).”
Membership Inference Attacks Against Machine Learning Models
Shokri, R., Stronati, M., Song, C., & Shmatikov, V. (2017) · 2017 IEEE Symposium on Security and Privacy (SP) · doi:10.1109/sp.2017.41 · Cited by 4,754 · cited 3×
“Securing complete privacy in FL remains questionable, as researchers have been able to develop successful attack models that leak PII (Hitaj et al., 2017; Shokri et al., 2017).”
BackgroundMethodThe 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 3×
“When two institutions are training a single ML model, there exists a tradeoff between the privacy of the data and the joint effort of model development (Rieke et al., 2020).”
BackgroundInfluentialPractical Secure Aggregation for Privacy-Preserving Machine Learning
Bonawitz, K., Ivanov, V., Kreuter, B., Marcedone, A., McMahan, H. B., Patel, S., Ramage, D., Segal, A., & Seth, K. (2017) · Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/3133956.3133982 · Cited by 3,956
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
Federated Learning With Differential Privacy: Algorithms and Performance Analysis
Wei, K., Li, J., Ding, M., Ma, C., Yang, H. H., Farokhi, F., Jin, S., Quek, T. Q. S., & Vincent Poor, H. (2020) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2020.2988575 · Cited by 2,400 · cited 2×
“K-random sampling of clients to participate in each model aggregation (Wei et al., 2020)”
BackgroundMethodPrivacy-preserving deep learning
Shokri, R., & Shmatikov, V. (2015) · Proceedings of the 22nd ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/2810103.2813687 · Cited by 2,356 · cited 1×
“Selective parameter sharing, in which a fraction of an asynchronous local update is communicated to the global server (Shokri and Shmatikov, 2015) 4.”
BackgroundPrivacy-Preserving Deep Learning via Additively Homomorphic Encryption
Phong, L. T., Aono, Y., Hayashi, T., Wang, L., & Moriai, S. (2018) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2017.2787987 · Cited by 1,662
Deep Models Under the GAN: Information Leakage from Collaborative Deep Learning
Hitaj, B., Ateniese, G., & Perez-Cruz, F. (2017) · Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/3133956.3134012 · Cited by 1,377 · cited 7×
“Such an attack might be used to de-identify patient data in a medical application of FL (Hitaj et al., 2017).”
BackgroundMethodInfluentialKahn, M. G., Callahan, T. J., Barnard, J., Bauck, A. E., Brown, J., Davidson, B. N., Estiri, H., Goerg, C., Holve, E., Johnson, S. G., et al. (2016) · eGEMs (Generating Evidence & Methods to improve patient outcomes) · doi:10.13063/2327-9214.1244 · Cited by 697 · cited 1×
“Functionally, Data Completeness is an important measure for the conformity of the model (Kahn et al., 2016).”
MethodPrivacy-Preserving Federated Brain Tumour Segmentation
Li, W., Milletarì, F., Xu, D., Rieke, N., Hancox, J., Zhu, W., Baust, M., Cheng, Y., Ourselin, S., Cardoso, M. J., et al. (2019) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-32692-0_16 · Cited by 578
Köchling, A., & Wehner, M. C. (2020) · Business Research · doi:10.1007/s40685-020-00134-w · Cited by 511 · cited 2×
“Data Heterogeneity: We sought to include Data Heterogeneity as one of the principal components for determining Data Contribution due to the signi fi cant legal and ethical issues of Algorithmic Discrimination present in ML.”
BackgroundLegal and human rights issues of AI: Gaps, challenges and vulnerabilities
Rodrigues, R. (2020) · Journal of Responsible Technology · doi:10.1016/j.jrt.2020.100005 · Cited by 494 · cited 1×
“…issues that are inherent in ML currently are: lack of transparency for algorithms, lack of contestability for non-optimal results, the legal status of AI with questions of “ personhood, ” and correspondingly, a lack of accountability for damage that may arise from the algorithm (Rodrigues, 2020).”
BackgroundSecure Logistic Regression Based on Homomorphic Encryption: Design and Evaluation
Kim, M., Song, Y., Wang, S., Xia, Y., & Jiang, X. (2018) · JMIR Medical Informatics · doi:10.2196/medinform.8805 · Cited by 257 · cited 1×
“Despite its potential bene fi t, the additional computational burden and having insuf fi cient supporting frameworks and libraries make homomorphic encryption not easily implementable (Kim et al., 2018).”
BackgroundBig data and black-box medical algorithms
Price, W. N. (2018) · Science Translational Medicine · doi:10.1126/scitranslmed.aao5333 · Cited by 254
Federated learning improves site performance in multicenter deep learning without data sharing
Sarma, K. V., Harmon, S., Sanford, T., Roth, H. R., Xu, Z., Tetreault, J., Xu, D., Flores, M. G., Raman, A. G., Kulkarni, R., et al. (2021) · Journal of the American Medical Informatics Association · doi:10.1093/jamia/ocaa341 · Cited by 242 · cited 1×
“Given the large volume and diverse data needed for model training, Federated Learning (FL) approaches may provide a novel opportunity for the future of ML applications (Rajendran et al., 2021; Sarma et al., 2021).”
BackgroundPrivacy-Preserving Patient Similarity Learning in a Federated Environment: Development and Analysis
Lee, J., Sun, J., Wang, F., Wang, S., Jun, C.-H., & Jiang, X. (2018) · JMIR Medical Informatics · doi:10.2196/medinform.7744 · Cited by 196
Duan, R., Boland, M. R., Liu, Z., Liu, Y., Chang, H. H., Xu, H., Chu, H., Schmid, C. H., Forrest, C. B., Holmes, J. H., et al. (2020) · Journal of the American Medical Informatics Association · doi:10.1093/jamia/ocz199 · Cited by 99
Cloud-based federated learning implementation across medical centers
Rajendran, S., Obeid, J. S., Binol, H., D`Agostino, R., Foley, K., Zhang, W., Austin, P., Brakefield, J., Gurcan, M. N., & Topaloglu, U. (2021) · JCO Clinical Cancer Informatics · doi:10.1200/cci.20.00060 · Cited by 76 · cited 1×
“Given the large volume and diverse data needed for model training, Federated Learning (FL) approaches may provide a novel opportunity for the future of ML applications (Rajendran et al., 2021; Sarma et al., 2021).”
BackgroundSelf-citationCan we trust AI not to further embed racial bias and prejudice?
Noor, P. (2020) · BMJ · doi:10.1136/bmj.m363 · Cited by 71 · cited 1×
“Even 6 years later, ML still presents problems in healthcare, with melanoma detection algorithms being primarily trained on light-skinned individuals; Black people, while less likely to develop melanoma, are more likely to die from it (Noor, 2020).”
BackgroundAllen, B., Agarwal, S., Kalpathy-Cramer, J., & Dreyer, K. (2019) · Journal of the American College of Radiology · doi:10.1016/j.jacr.2019.04.023 · Cited by 40
Cited by
Guidelines and evaluation of clinical explainable AI in medical image analysis
Jin, W., Li, X., Fatehi, M., & Hamarneh, G. (2023) · Medical Image Analysis · doi:10.1016/j.media.2022.102684 · Cited by 199
Myrzashova, R., Alsamhi, S. H., Shvetsov, A. V., Hawbani, A., & Wei, X. (2023) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3263598 · Cited by 178
“to store and process sensitive patient information [4], [5], [6]”
BackgroundRajendran, S., Pan, W., Sabuncu, M. R., Chen, Y., Zhou, J., & Wang, F. (2024) · Patterns · doi:10.1016/j.patter.2023.100913 · Cited by 47
Self-citation
Metadata
- Authors
- Topaloglu, M. Y., Morrell, E. M., Rajendran, S., & Topaloglu, U.
- Venue
- Frontiers in Artificial Intelligence
- Publisher
- Frontiers Media SA
- Published
- 2021-10-06
- Volume
- 4
- Pages
- 746497-746497
- ISSN
- 2624-8212 (ISSN-L)
- OpenAlex
- W3202840499
- Cited by
- 50
- References
- 22
- FWCI
- 2.29
- Type
- journal article
- Language
- en
- Open access
- gold
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.