Skip to content

Homomorphic Encryption for Machine Learning in Medicine and Bioinformatics

Authors: Alexander Wood; Kayvan Najarian; Delaram Kahrobaei
Year: 2021
Venue: ACM Computing Surveys

type: survey collection: ppml status: auto

Links: DOI

Summary

Auto-extracted summary (unreviewed)

This ACM Computing Surveys article surveys fully homomorphic encryption and its use in medicine and bioinformatics. It introduces FHE concepts and history, organizes current schemes into Boolean-circuit (FHEW/TFHE), exact-arithmetic (BGV/BFV) and approximate-arithmetic (CKKS) models, catalogs open-source libraries such as SEAL, HElib, TFHE and HEAAN, and reviews privacy-preserving machine learning on medical data (deep learning, decision trees, Naive Bayes, encrypted training) as well as secure genomic algorithms such as paternity/ancestry testing and GWAS.

Abstract

Machine learning and statistical techniques are powerful tools for analyzing large amounts of medical and genomic data. On the other hand, ethical concerns and privacy regulations prevent free sharing of this data. Encryption techniques such as fully homomorphic encryption (FHE) enable evaluation over encrypted data. Using FHE, machine learning models such as deep learning, decision trees, and Naive Bayes have been implemented for privacy-preserving applications using medical data. These applications include classifying encrypted data and training models on encrypted data. FHE has also been shown to enable secure genomic algorithms, such as paternity and ancestry testing and privacy-preserving applications of genome-wide association studies. This survey provides an overview of fully homomorphic encryption and its applications in medicine and bioinformatics. The high-level concepts behind FHE and its history are introduced, and details on current open-source implementations are provided. The state of fully homomorphic encryption for privacy-preserving techniques in machine learning and bioinformatics is reviewed, along with descriptions of how these methods can be implemented in the encrypted domain.
Source: crossref

Extracted details

Field Value Source Confidence
contribution overview of FHE concepts, history and the three models of homomorphic computation llm 0.85
contribution comparison of open-source FHE libraries and tools llm 0.85
contribution review of FHE-based machine learning on medical data (classification and training) llm 0.85
contribution review of secure genomic algorithms and privacy-preserving GWAS llm 0.85
schemes BGV llm 0.85
schemes BFV llm 0.85
schemes FHEW llm 0.85
schemes TFHE llm 0.85
schemes CKKS llm 0.85
schemes GSW llm 0.85
libraries SEAL llm 0.85
libraries HElib llm 0.85
libraries TFHE llm 0.85
libraries HEAAN llm 0.85
models neural networks llm 0.85
models decision trees llm 0.85
models Naive Bayes llm 0.85
tasks medical data classification llm 0.85
tasks training models on encrypted data llm 0.85
tasks genomic algorithms (paternity and ancestry testing) llm 0.85
tasks genome-wide association studies llm 0.85
interaction non-interactive FHE llm 0.85
code url https://github.com/Microsoft/SEAL.4https (URL found in the full text) regex 0.50
code url https://github.com/homenc/HElib.6https (URL found in the full text) regex 0.50
code url https://github.com/vernamlab/cuFHE.8https (URL found in the full text) regex 0.50
llm relevance low llm 0.85

Keyword mentions

Counts of vocabulary terms in the full text (first page of occurrence in brackets).

  • scheme: CKKS 37 [p.8], Leveled / somewhat HE 23 [p.4], TFHE 21 [p.8], BFV 20 [p.6], BGV 20 [p.6], FHEW 11 [p.8], NTRU 11 [p.5], GSW 10 [p.7], Paillier 8 [p.4], Oblivious transfer 8 [p.17], Partially homomorphic encryption 8 [p.4], LTV 7 [p.7], MPC 7 [p.12], Garbled circuits 5 [p.12], DGHV 4 [p.7], RSA 4 [p.4], YASHE 3 [p.7], Secret sharing 3 [p.12], ElGamal 2 [p.4], BGN 2 [p.4], Differential privacy 2 [p.12]
  • library: Microsoft SEAL 18 [p.9], HElib 18 [p.9], PALISADE 11 [p.9], Pyfhel 4 [p.11], cuFHE / nuFHE 4 [p.10], Lattigo 3 [p.9], nGraph-HE 2 [p.9], cuHE 1 [p.11], ABY / ABY3 1 [p.34]
  • model: Logistic regression 26 [p.3], Decision tree 21 [p.2], Naive Bayes 13 [p.2], k-means 4 [p.21], CNN 1 [p.18], SVM 1 [p.12]
  • dataset: Heart disease 2 [p.14], GLUE tasks (QNLI/RTE/MRPC/CoLA/QQP/MNLI/STS-B) 1 [p.9], Breast cancer (Wisconsin) 1 [p.14]
  • hardware: CUDA 5 [p.10], GPU 4 [p.6]
  • technique: Bootstrapping 36 [p.5], ReLU 15 [p.19], Sigmoid 12 [p.13], Convolution 11 [p.18], Compiler 9 [p.9], RNS 8 [p.8], Polynomial approximation 6 [p.8], Square activation 6 [p.20], SIMD packing 6 [p.6], Modulus switching 5 [p.6], Chebyshev 2 [p.13], Taylor expansion 2 [p.13], Pruning 2 [p.16], Non-interactive 2 [p.21]
  • assumption: Lattice-based 26 [p.3], LWE 22 [p.5], RLWE 5 [p.5]
  • task: Medical / healthcare 63 [p.2], Genomics 62 [p.2], Clustering 14 [p.12], Finance 2 [p.29]
  • topic: Homomorphic encryption 212 [p.2], Open source 35 [p.2], Privacy-preserving machine learning 4 [p.3]

Files

Path (under papers/) Collection Pages Size SHA-256 Note
references/ppml/surveys/wood-2020_homomorphic-encryption-for-ml-in-medicine-and-bioinformatics.pdf ppml/surveys 36 1002.8 KB 9660c4150ab4 was Wood and Najarian - 2020 - Homomorphic Encryption for Machine Learning in Medicine and Bioinformatics.pdf

Database key: homomorphic-encryption-for-machine-learning-in-medicine-and · created 2026-11-05 · updated 2026-11-05 · edit with python3 tools/paperdb.py set homomorphic-encryption-for-machine-learning-in-medicine-and COLUMN VALUE