EUVIMEDEUVIMED entdecken
Open Access

PPML-Omics: A privacy-preserving federated machine learning method protects patients’ privacy in omic data

Juexiao Zhou iD, Siyuan Chen iD, Yulian Wu, Haoyang Li iD, Bin Zhang, Longxi Zhou iD, Yan Hu, Zihang Xiang iD, Zhongxiao Li iD, Ningning Chen iD, Wenkai Han iD, Chencheng Xu iD und 2 weitere

Science Advances · 02.02.2024 · 10 (5)

DOI: 10.1126/sciadv.adh8601 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Science Advances, erschienen am 02.02.2024. Verfasst von Juexiao Zhou und 13 weiteren Autor:innen. Beteiligte Einrichtung: Computer Science Program, Computer, Electrical and Mathematical Sciences and Engineering Division, King Abdullah University of Science and Technology…. Hauptthemen laut MEDLINE: Privacy, Algorithms. Die Arbeit wurde bisher 64-mal zitiert (Crossref) und nennt 109 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Modern machine learning models toward various tasks with omic data analysis give rise to threats of privacy leakage of patients involved in those datasets. Here, we proposed a secure and privacy-preserving machine learning method (PPML-Omics) by designing a decentralized differential private federated learning algorithm. We applied PPML-Omics to analyze data from three sequencing technologies and addressed the privacy concern in three major tasks of omic data under three representative deep learning models. We examined privacy breaches in depth through privacy attack experiments and demonstrated that PPML-Omics could protect patients’ privacy. In each of these applications, PPML-Omics was able to outperform methods of comparison under the same level of privacy guarantee, demonstrating the versatility of the method in simultaneously balancing the privacy-preserving capability and utility in omic data analysis. Furthermore, we gave the theoretical proof of the privacy-preserving capability of PPML-Omics, suggesting the first mathematically guaranteed method with robust and generalizable empirical performance in protecting patients’ privacy in omic data.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

PrivacyAlgorithmsHumansData AnalysisMachine LearningTechnology

Autor:innen und Institutionen

  • Computer Science Program, Computer, Electrical and Mathematical Sciences and Engineering Division, King Abdullah University of Science and Technology (KAUST), Thuwal 23955-6900, Kingdom of Saudi Arabia.
  • Computational Bioscience Research Center, Computer, Electrical and Mathematical Sciences and Engineering Division, King Abdullah University of Science and Technology (KAUST), Thuwal 23955-6900, Kingdom of Saudi Arabia.

ORCID-Profile: Juexiao Zhou, Siyuan Chen, Haoyang Li, Longxi Zhou, Zihang Xiang, Zhongxiao Li, Ningning Chen, Wenkai Han, Chencheng Xu, Di Wang, Xin Gao

Zitieren

APA

Zhou, J., Chen, S., Wu, Y., Li, H., Zhang, B., Zhou, L., Hu, Y., Xiang, Z., Li, Z., Chen, N., Han, W., Xu, C., Wang, D., & Gao, X. (2024). PPML-Omics: A privacy-preserving federated machine learning method protects patients’ privacy in omic data. Science Advances, 10(5). https://doi.org/10.1126/sciadv.adh8601

Vancouver

Zhou J, Chen S, Wu Y, Li H, Zhang B, Zhou L, et al. PPML-Omics: A privacy-preserving federated machine learning method protects patients’ privacy in omic data. Science Advances. 2024;10(5). doi:10.1126/sciadv.adh8601

BibTeX
@article{Zhou2024,
  author = {Zhou, Juexiao and Chen, Siyuan and Wu, Yulian and Li, Haoyang and Zhang, Bin and Zhou, Longxi and Hu, Yan and Xiang, Zihang and Li, Zhongxiao and Chen, Ningning and Han, Wenkai and Xu, Chencheng and Wang, Di and Gao, Xin},
  title = {PPML-Omics: A privacy-preserving federated machine learning method protects patients’ privacy in omic data},
  journal = {Science Advances},
  year = {2024},
  volume = {10},
  number = {5},
  doi = {10.1126/sciadv.adh8601}
}
Kostenlos starten →