EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Aim high, stay private: differentially private synthetic data enables public release of behavioral health information with high utility

Mohsen Ghasemizade iD, Juniper Lovato, Chris Danforth, Peter Sheridan Dodds, Laura S P Bloomfield, Matthew Price, Joseph Near

JAMIA Open · 06.03.2026 · 9 (2)

DOI: 10.1093/jamiaopen/ooag066 ↗

Kurz gefasst

Fachartikel in der Zeitschrift JAMIA Open, online erschienen am 21.04.2026. Verfasst von Mohsen Ghasemizade und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science, University of Vermont, VT, 05405, United States. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: National Science Foundation, MassMutual Center of Excellence in Complex Systems and Data Science.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objective

Sharing behavioral health and wearable data poses privacy challenges, as traditional de-identification remains vulnerable to re-identification. Differential privacy (DP) provides mathematical guarantees through a tunable privacy budget, ϵ. This study evaluates the feasibility of generating and releasing DP synthetic behavioral health data with high analytical utility, identifying practical ϵ values for public data sharing. Materials and methods We analyzed physiological data from wearable devices and self-reported data from Phase 1 of the Lived Experiences Measured Using Rings Study (LEMURS), which tracked sleep, stress, and well-being among first-year college students. Three DP synthetic data generators: AIM, MST, and PATECTGAN, were evaluated across privacy budgets ranging from ϵ=1 to 100. Utility was assessed using L1/L2 errors, correlation, regression, UMAP, and assessed vulnerability via privacy attacks.

Results

AIM outperformed MST and PATECTGAN in preserving both statistical and analytical properties of the original data. For the Survey dataset, the lowest marginal errors occurred at ϵ=5 and 10. Correlation, regression, and UMAP analyses confirmed that AIM-generated data closely replicated original relationships at moderate ϵ values.

Discussion

Choice of privacy budget is still an open question, and it is task-agnostic and dataset-specific. Moderate privacy budgets (5≤ϵ≤10) maintained key associations between physiological and psychological measures while ensuring privacy. AIM’s workload-aware design effectively allocated noise toward relevant features, enhancing performance.

Conclusion

A privacy budget of ϵ=5 offers a practical balance between data utility and participant privacy for LEMURS behavioral health data sharing.

Schlagwörter

Stichwörter der Autor:innen

differential privacyhealth behavioral studyutility-privacy tradeoffwearable devices

Autor:innen und Institutionen

  • Department of Computer Science, University of Vermont, VT, 05405, United States.
  • Department of Mathematics and Statistics, University of Vermont, VT, 05405, United States.
  • The Gund Institute for the Environment, University of Vermont, VT, 05405, United States.
  • Department of Psychological Science, University of Vermont, VT, 05405, United States.

ORCID-Profile: Mohsen Ghasemizade

Förderung

  • National Science Foundation
  • MassMutual Center of Excellence in Complex Systems and Data Science

Zitieren

APA

Ghasemizade, M., Lovato, J., Danforth, C., Dodds, P. S., Bloomfield, L. S. P., Price, M., & Near, J. (2026). Aim high, stay private: differentially private synthetic data enables public release of behavioral health information with high utility. JAMIA Open, 9(2). https://doi.org/10.1093/jamiaopen/ooag066

Vancouver

Ghasemizade M, Lovato J, Danforth C, Dodds PS, Bloomfield LSP, Price M, et al. Aim high, stay private: differentially private synthetic data enables public release of behavioral health information with high utility. JAMIA Open. 2026;9(2). doi:10.1093/jamiaopen/ooag066

BibTeX
@article{Ghasemizade2026,
  author = {Ghasemizade, Mohsen and Lovato, Juniper and Danforth, Chris and Dodds, Peter Sheridan and Bloomfield, Laura S P and Price, Matthew and Near, Joseph},
  title = {Aim high, stay private: differentially private synthetic data enables public release of behavioral health information with high utility},
  journal = {JAMIA Open},
  year = {2026},
  volume = {9},
  number = {2},
  doi = {10.1093/jamiaopen/ooag066}
}
Kostenlos starten →