Fair compute loads enabled by blockchain: sharing models by alternating client and server roles.

Journal: Journal of the American Medical Informatics Association : JAMIA
Published Date:

Abstract

OBJECTIVE: Decentralized privacy-preserving predictive modeling enables multiple institutions to learn a more generalizable model on healthcare or genomic data by sharing the partially trained models instead of patient-level data, while avoiding risks such as single point of control. State-of-the-art blockchain-based methods remove the "server" role but can be less accurate than models that rely on a server. Therefore, we aim at developing a general model sharing framework to preserve predictive correctness, mitigate the risks of a centralized architecture, and compute the models in a fair way.

Authors

  • Tsung-Ting Kuo
    University of California San Diego, La Jolla, CA.
  • Rodney A Gabriel
    Department of Medicine, Division of Biomedical Informatics, University of California, San Diego, La Jolla, CA, USA.
  • Lucila Ohno-Machado
    University of California San Diego, La Jolla, CA.