Privacy-preserving semi-parallel logistic regression training with fully homomorphic encryption

Abstract

Background: Privacy-preserving computations on genomic data, and more generally on medical data, is a critical path technology for innovative, life-saving research to positively and equally impact the global population. It enables medical research algorithms to be securely deployed in the cloud because operations on encrypted genomic databases are conducted without revealing any individual genomes. Methods for secure computation have shown significant performance improvements over the last several years. However, it is still challenging to apply them on large biomedical datasets. Methods: The HE Track of iDash 2018 competition focused on solving an important problem in practical machine learning scenarios, where a data analyst that has trained a regression model (both linear and logistic) with a certain set of features, attempts to find all features in an encrypted database that will improve the quality of the model. Our solution is based on the hybrid framework Chimera that allows for switching between different families of fully homomorphic schemes, namely TFHE and HEAAN. Results: Our solution is one of the finalist of Track 2 of iDash 2018 competition. Among the submitted solutions, ours is the only bootstrapped approach that can be applied for different sets of parameters without re-encrypting the genomic database, making it practical for real-world applications. Conclusions: This is the first step towards the more general feature selection problem across large encrypted databases.

iDASH secure genome analysis competition 2018: blockchain genomic data access logging, homomorphic encryption on GWAS, and DNA segment searching

“…We received 13 solutions, submitted from 7 teams by the deadline. The research papers from the participating teams to describe the methods are included in this special issue [26][27][28][29][30]. to obtain mean and standard deviation.…”

CHIMERA: Combining Ring-LWE-based Fully Homomorphic Encryption Schemes

“…Some of the algorithms presented in this paper have been implemented and tested in the context of an IDash 2018 [1] submission described in [9]. This submission was selected in October 2018 to be among the finalists of the competition.…”

“…Other directions that we leave as a future work is the elaboration of a bridge towards the BGV scheme as well as the introduction of RNS in this framework. The presented framework has already been applied to some concrete use-cases in the domain of machine-learning [5,9] but we are wishing to provide more applications in this or other directions. The remaining bridges will be implemented when specific use cases are identified for them.…”

“…We then evaluate the Taylor expansion of the complex exponential (up to degree d = E ← HeaanMult(E, E) (the new E has parameters τ = 0 and level L + (p − j)ρ) 9: end for 10: return E at level L times the multiplicative depth, so the level of E is ≤ L + pρ. Finally, we square the ciphertext p times to obtain the desired result.…”

Homomorphic Encryption for Machine Learning in Medicine and Bioinformatics

“…Blatt et al approximate the sigmoid function with Chebyshev polynomials for an implementation of logistic regression in PALISADE along with a custom variant of the CKKS scheme [29]. Carpov et al perform logistic regression over data encrypted using TFHE and HEAAN [48]. Additional methods use HElib [201] and gradient descent with a minimax approximation for the sigmoid in SEAL [53].…”