Privacy and utility of private synthetic data for medical data analyses
Open accessAbstract
The increasing availability and use of sensitive personal data raises a set of issues regarding the privacy of the individuals behind the data. These concerns become even more important when health data are processed, as are considered sensitive (according to most global regulations). Privacy Enhancing Technologies (PETs) attempt to protect the privacy of individuals whilst preserving the utility of data. One of the most popular technologies recently is Differential Privacy (DP), which was used for the 2020 U.S. Census. Another trend is to combine synthetic data generators with DP to create so-called private synthetic data generators. The objective is to preserve statistical properties as accurately as possible, while the generated data should be as different as possible compared to the original data regarding private features. While these technologies seem promising, there is a gap between academic research on DP and synthetic data and the practical application and evaluation of these techniques for real-world use cases. In this paper, we evaluate three different private synthetic data generators (MWEM, DP-CTGAN, and PATE-CTGAN) on their use-case-specific privacy and utility. For the use case, continuous heart rate measurements from different individuals are analyzed. This work shows that private synthetic data generators have tremendous advantages over traditional techniques, but also require in-depth analysis depending on the use case. Furthermore, it can be seen that each technology has different strengths, so there is no clear winner. However, DP-CTGAN often performs slightly better than the other technologies, so it can be recommended for a continuous medical data use case.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
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?
Synthetic data generators were applied for building similar replicas of original private datasets while maintaining consistent statistical properties.
“Additionally, the synthetic data generators were applied for building similar replicas of the original private datasets, while maintaining consistent statistical properties [161,162].”
From Challenges in AI-driven Biomedical Multimodal Data Fusion and Analysis (2025) · cites this paper as [161]
Cites
Generative adversarial networks
Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., & Bengio, Y. (2020) · Communications of the ACM · doi:10.1145/3422622 · Cited by 18,384 · see citing works
k-ANONYMITY: A MODEL FOR PROTECTING PRIVACY
SWEENEY, L. (2002) · International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems · doi:10.1142/s0218488502001648 · Cited by 8,571 · see citing works
Calibrating Noise to Sensitivity in Private Data Analysis
Dwork, C., McSherry, F., Nissim, K., & Smith, A. (2006) · Lecture Notes in Computer Science · doi:10.1007/11681878_14 · Cited by 7,272 · see citing works
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 · see citing works
<i>L</i> -diversity: Privacy beyond <i>k</i> -anonymity
Machanavajjhala, A., Kifer, D., Gehrke, J., & Venkitasubramaniam, M. (2007) · ACM Transactions on Knowledge Discovery from Data · doi:10.1145/1217299.1217302 · Cited by 3,617 · see citing works
t-Closeness: Privacy Beyond k-Anonymity and l-Diversity
Li, N., Li, T., & Venkatasubramanian, S. (2007) · 2007 IEEE 23rd International Conference on Data Engineering · doi:10.1109/icde.2007.367856 · Cited by 3,451 · see citing works
Privacy-preserving data publishing: A survey of recent developments
Fung, B. C. M., Wang, K., Chen, R., & Yu, P. S. (2010) · ACM Computing Surveys · doi:10.1145/1749603.1749605 · Cited by 1,704 · cited 1× · see citing works
“More sophisticated technologies such as Differential Privacy (DP) provide a statistical measure that quantifies the privacy of the individuals contained in the dataset [4].”
BackgroundUnique in the Crowd: The privacy bounds of human mobility
de Montjoye, Y.-A., Hidalgo, C. A., Verleysen, M., & Blondel, V. D. (2013) · Scientific Reports · doi:10.1038/srep01376 · Cited by 1,645 · see citing works
Modeling Tabular data using Conditional GANPreprint
Xu, L., Skoularidou, M., Cuesta‐Infante, A., & Veeramachaneni, K. K. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1907.00503 · Cited by 983 · see citing works
Synthetic data in machine learning for medicine and healthcare
Chen, R. J., Lu, M. Y., Chen, T. Y., Williamson, D. F. K., & Mahmood, F. (2021) · Nature Biomedical Engineering · doi:10.1038/s41551-021-00751-8 · Cited by 728 · see citing works
Scalable Private Learning with PATE
Papernot, N., Song, S., Mironov, I., Raghunathan, A., Talwar, K., & Erlingsson, Ú. (2018) · arXiv · doi:10.48550/arxiv.1802.08908 · Cited by 373 · see citing works
Hiding the presence of individuals from shared databases
Nergiz, M. E., Atzori, M., & Clifton, C. (2007) · Proceedings - ACM-SIGMOD International Conference on Management of Data · doi:10.1145/1247480.1247554 · Cited by 307 · see citing works
How To Break Anonymity of the Netflix Prize DatasetPreprint
Narayanan, A., & Shmatikov, V. (2006) · arXiv (Cornell University) · doi:10.48550/arxiv.cs/0610105 · Cited by 275 · see citing works
Statistical Disclosure Limitation
Cox, L. H. (2014) · Wiley StatsRef: Statistics Reference Online · doi:10.1002/9781118445112.stat00072 · Cited by 147 · see citing works
DataSynthesizer: Privacy-Preserving Synthetic Datasets
Ping, H., Stoyanovich, J., & Howe, B. (2017) · Proceedings of the 29th International Conference on Scientific and Statistical Database Management · doi:10.1145/3085504.3091117 · Cited by 144 · see citing works
Differential Privacy: On the Trade-Off between Utility and Information Leakage
Alvim, M. S., Andrés, M. E., Konstantinos Chatzikokolakis, A., Degano, P., & Catuscia Palamidessi, A. (2012) · Lecture notes in computer science · doi:10.1007/978-3-642-29420-4_3 · Cited by 123 · see citing works
Synthetic Data -- Anonymisation Groundhog DayPreprint
Stadler, T., Oprisanu, B., & Troncoso, C. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.07018 · Cited by 115 · see citing works
Privacy and Synthetic DatasetsPreprint
Bellovin, S. M., Dutta, P. K., & Reitinger, N. (2018) · SSRN Electronic Journal · doi:10.2139/ssrn.3255766 · Cited by 96 · see citing works
Machine Learning and Knowledge Discovery in Databases
(2019) · Lecture notes in computer science · doi:10.1007/978-3-030-10928-8 · Cited by 86 · see citing works
Secondary Use and Analysis of Big Data Collected for Patient Care
Martin-Sanchez, F. J., Aguiar‐Pulido, V., López–Campos, G., Peek, N. B., & Sacchi, L. (2017) · Yearbook of Medical Informatics · doi:10.15265/iy-2017-008 · Cited by 67 · see citing works
Bowen, C., & Snoke, J. (2019) · Journal of Privacy and Confidentiality · doi:10.29012/jpc.748 · Cited by 64 · see citing works
Differentially Private Synthetic Data: Applied Evaluations and EnhancementsPreprint
Rosenblatt, L., Liu, X., Pouyanfar, S., de Leon, E., Desai, A., & Allen, J. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.05537 · Cited by 45 · see citing works
Statistical Disclosure Limitation
Cox, L. H. (2005) · Encyclopedia of Statistical Sciences · doi:10.1002/0471667196.ess1166.pub2 · Cited by 43 · see citing works
Browsing Unicity: On the Limits of Anonymizing Web Tracking Data
Deusser, C., Passmann, S., & Strufe, T. (2020) · IEEE Symposium on Security and Privacy (S&P) · doi:10.1109/sp40000.2020.00018 · Cited by 28 · see citing works
Differential Privacy and the 2020 US Census
Garfinkel, S. (2022) · doi:10.21428/2c646de5.7ec6ab93 · Cited by 21 · see citing works
Re-Identification Risk in HIPAA De-Identified Datasets: The MVA Attack.
Janmey, V., & Elkin, P. L. (2018) · Europe PMC (PubMed Central) · Cited by 19 · see citing works
Differentially Private Medical Texts Generation Using Generative Neural Networks
Aziz, M. M. A., Ahmed, T., Faequa, T., Jiang, X., Yao, Y. Y., & Mohammed, N. (2021) · ACM Transactions on Computing for Healthcare · doi:10.1145/3469035 · Cited by 18 · see citing works
Evaluating Variational Autoencoder as a Private Data Release Mechanism for Tabular Data
Li, S.-C., Tai, B.-C., & Huang, Y. (2019) · IEEE Pacific Rim International Symposium on Dependable Computing (PRDC) · doi:10.1109/prdc47002.2019.00050 · Cited by 18 · see citing works
Flagged first, then by citations
3 of the 31 references couldn’t be matched to a known paper.
Cited by
Sorted by citations
Harnessing the power of synthetic data in healthcare: innovation, application, and privacy
Giuffrè, M., & Shung, D. L. (2023) · npj Digital Medicine · doi:10.1038/s41746-023-00927-3 · Cited by 392 · see citing works
Synthetic data generation methods in healthcare: A review on open-source tools and methods
Pezoulas, V. C., Zaridis, D. I., Mylona, E., Androutsos, C., Apostolidis, K., Tachos, N. S., & Fotiadis, D. I. (2024) · Computational and Structural Biotechnology Journal · doi:10.1016/j.csbj.2024.07.005 · Cited by 270 · see citing works
A scoping review of privacy and utility metrics in medical synthetic data
Kaabachi, B., Despraz, J., Meurers, T., Otte, K., Halilovic, M., Kulynych, B., Praßer, F., & Raisaro, J. L. (2025) · npj Digital Medicine · doi:10.1038/s41746-024-01359-3 · Cited by 108 · see citing works
Challenges in AI-driven Biomedical Multimodal Data Fusion and Analysis
Liu, J., Cen, X., Yi, C., Wang, F.-A., Ding, J., Cheng, J., Wu, Q., Gai, B., Zhou, Y., He, R., et al. (2025) · Genomics, Proteomics & Bioinformatics · doi:10.1093/gpbjnl/qzaf011 · Cited by 75 · see citing works
Vallevik, V. B., Babić, A., Marshall, S., Elvatun, S., Brøgger, H. M. B., Alagaratnam, S., Edwin, B., Veeraragavan, N. R., Befring, A. K., & Nygård, J. F. (2024) · International Journal of Medical Informatics · doi:10.1016/j.ijmedinf.2024.105413 · Cited by 61 · see citing works
Alabdulwahab, S., Kim, Y.-T., Seo, A., & Son, Y. (2023) · Applied Sciences · doi:10.3390/app131910951 · Cited by 43 · see citing works
Privacy Mechanisms and Evaluation Metrics for Synthetic Data Generation: A Systematic Review
Osorio-Marulanda, P. A., Epelde, G., Hernandez, M., Isasa, I., Reyes, N. M., & Beristain, A. (2024) · IEEE Access · doi:10.1109/access.2024.3417608 · Cited by 35 · see citing works
Synthetic data in medicine: Legal and ethical considerations for patient profiling
Nišević, M., Milojevic, D., & Spajić, D. (2025) · Computational and Structural Biotechnology Journal · doi:10.1016/j.csbj.2025.05.026 · Cited by 32 · see citing works
Systematic Review of Generative Modelling Tools and Utility Metrics for Fully Synthetic Tabular Data
Lautrup, A. D., Hyrup, T., Zimek, A., & Schneider–Kamp, P. (2024) · ACM Computing Surveys · doi:10.1145/3704437 · Cited by 26 · see citing works
Advancing privacy in learning analytics using differential privacy
Liu, Q., Shakya, R., Khalil, M., & Jovanović, J. (2025) · International Conference on Learning Analytics and Knowledge (LAK) · doi:10.1145/3706468.3706493 · Cited by 30 · see citing works
Ensuring privacy through synthetic data generation in education
Liu, Q., Shakya, R., Jovanović, J., Khalil, M., & de la Hoz‐Ruiz, J. (2025) · British Journal of Educational Technology · doi:10.1111/bjet.13576 · Cited by 21 · see citing works
Zhou, Z., Guo, Y.-C., Tang, R., Liang, H., He, J., & Xu, F. (2024) · npj Digital Medicine · doi:10.1038/s41746-024-01290-7 · Cited by 18 · see citing works
Majeed, A., & Hwang, S. O. (2024) · IEEE Transactions on Big Data · doi:10.1109/tbdata.2024.3442534 · Cited by 12 · see citing works
High-Fidelity Synthetic Data Applications for Data Augmentation
Wang, Z., Draghi, B., Rotalinti, Y., Lunn, D. R., & Myles, P. R. (2024) · Artificial intelligence · doi:10.5772/intechopen.113884 · Cited by 11 · see citing works
GAN-Based Novel Approach for Generating Synthetic Medical Tabular Data
Nasimov, R., Nasimova, N., Mirzakhalilov, S., Tokdemir, G., Rizwan, M., Abdusalomov, A. B., & Cho, Y. I. (2024) · Bioengineering · doi:10.3390/bioengineering11121288 · Cited by 8 · see citing works
Enhanced Dataset Synthesis Using CTGAN for Metagenomic Dataset
Ince, V., Bader-El-Den, M. B., & Sari, O. F. (2024) · IEEE International Conference on Intelligent Systems · doi:10.1109/is61756.2024.10705275 · Cited by 8 · see citing works
Kuo, N. I.-H., Gallego, B., & Jorm, L. R. (2025) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2025.104928 · Cited by 6 · see citing works
Graph Learning and Deep Neural Network Ensemble for Supporting Cognitive Decline Assessment
Antonesi, G., Rancea, A., Cioara, T., & Anghel, I. M. (2023) · Technologies · doi:10.3390/technologies12010003 · Cited by 4 · see citing works
Synthetic Data: Generate Avatar Data on Demand
Lebrun, T., Béziaud, L., Allard, T., Boutet, A., Gambs, S., & Maouche, M. (2024) · Lecture notes in computer science · doi:10.1007/978-981-96-0576-7_15 · Cited by 4 · see citing works
Yang, A. C., Pan, M.-L., Lu, H. H., Lien, C.-Y., Wang, D.-W., Chen, C., Tarng, D., Niu, D., Chiou, S.-H., Wu, C.-Y., et al. (2025) · Journal of Medical Internet Research · doi:10.2196/58954 · Cited by 4 · see citing works
Distributed Online Multi-Label Learning with Privacy Protection in Internet of Things
Huang, F., Yang, N., Chen, H., Bao, W., & Yuan, D. (2023) · Applied Sciences · doi:10.3390/app13042713 · Cited by 2 · see citing works
Deep Learning - Recent Findings and Research
Domínguez-Morales, M., Civit-Masot, J., Munoz-Saavedra, L., & Damaševičius, R. (2023) · Artificial intelligence · doi:10.5772/intechopen.107725 · Cited by 3 · see citing works
Enhancing Data Governance through Data-Centric AI: Case Study in New Zealand Government Sector
Wang, A. X., Simpson, C. R., & Nguyen, B. P. (2024) · International Conference on Computer and Automation Engineering · doi:10.1109/iccae59995.2024.10569739 · Cited by 2 · see citing works
Reproduction of Original Glioblastoma and Brain Metastasis Research Findings Using Synthetic Data
Davalan, W., Khalaf, R., & Diaz, R. J. (2025) · World Neurosurgery · doi:10.1016/j.wneu.2025.123808 · Cited by 2 · see citing works
Sun, J. (2025) · International Scientific Technical and Economic Research · doi:10.71451/istaer2539 · Cited by 1 · see citing works
Krüsselmann, K., Achterberg, J., Haas, M., Spruit, M. R., & Liem, M. C. A. (2026) · European Journal of Criminology · doi:10.1177/14773708261469812 · Cited by 1 · see citing works
Yang, A. C., Pan, M.-L., Lu, H. H., Lien, C.-Y., Wang, D., Chen, C., Tarng, D., Niu, D., Chiou, S.-H., Wu, C., et al. (2024) · doi:10.2196/preprints.58954 · Cited by 1 · see citing works
Cyberdefense Powered by Generative AI: A Comprehensive State –of- the -Art Review
Alsamman, A. A., & Dabagh, N. B. A. (2026) · Alkadhim journal for computer science. · doi:10.61710/9jq4y549
Benz, S., & Städter, T. (2025) · Springer eBooks · doi:10.1007/978-3-658-49252-6_4
Natural Language Processing Approaches for Closed-Domain Technical Dialogues
Boukhatem, N. M. (2024) · theses.fr (ABES) · doi:10.70675/15a0eb24z21caz4cfaz836cz86f80dfe8ac2
Garcia, L. E. C., Zambrano, J. R. L., & Rendón, A. D. Z. (2026) · Revista Científica de Informática ENCRIPTAR · doi:10.56124/encriptar.v9i17.011
Privacy-Preserving Federated Learning for Fraud Detection in Distributed Banking Networks
Uzoatu, I. B., Jogunola, O., Abdullahi, A. D., Adebisi, B., & Dargahi, T. (2026) · IEEE Internet of Things Journal · doi:10.1109/jiot.2026.3697575
A Machine Learning Framework for Autonomous Bilge Water Management
Dutt, S., & Kanungo, S. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-18132-9_5
Challenges of using AI-based synthetic data in health economics and outcomes research
Saha, A., Yue, Z., & Ngorsuraches, S. (2026) · Expert Review of Pharmacoeconomics & Outcomes Research · doi:10.1080/14737167.2026.2691184
Privacy-Preserving Generation of Synthetic Pathology Reports for Information Extraction
Lorenzo, A., Coulet, A., & Gardent, C. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-30710-1_8
Hassan, R., & Fathy, D. (2026) · Journal of Artificial Intelligence for Healthcare Systems · doi:10.68159/r206901001
- Authors
- Appenzeller, A., Leitner, M., Philipp, P., Krempel, E., & Beyerer, J.
- Affiliations
- Karlsruhe Institute of Technology (DE)
- Fraunhofer Institute of Optronics, System Technologies and Image Exploitation (DE)
- Munich University of Applied Sciences (DE)
- Corresponding author
- Appenzeller, Arno
- Venue
- Applied Sciences
- Publisher
- MDPI AG
- Published
- 2022-12-01
- Volume
- 12
- Issue
- 23
- Pages
- 12320
- ISSN
- 2076-3417 (ISSN-L)
- OpenAlex
- W4310539682
- Cited by
- 37
- References
- 31
- Citations by year
- 2026: 8 · 2025: 10 · 2024: 14 · 2023: 5
- FWCI
- 2.39
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Applied Sciences (publishedVersion, cc-by)
- Repository KITopen (Karlsruhe Institute of Technology) (submittedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Fraunhofer-Publica (Fraunhofer-Gesellschaft) (submittedVersion)
- KITopen (cc-by)
- Funders
- Fraunhofer Lighthouse Project
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- Physical Sciences › Computer Science › Information Systems › Cloud Data Security Solutions
- Social Sciences › Sociology and Political Science › Privacy, Security, and Data Protection
- Keywords
- differential privacy, privacy-utility trade-off, health data privacy, CTGAN, privacy-enhancing technologies, medical data analysis
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
