PPFLHE: A privacy-preserving federated learning scheme with homomorphic encryption for healthcare data¶
Authors: Bo Wang; Hongtao Li; Yina Guo; Jie Wang
Year: 2023
Venue: Applied Soft Computing
type: system collection: ppml status: auto
Links: DOI
Summary¶
Auto-extracted summary (unreviewed)
PPFLHE is a federated learning scheme for healthcare data in which clients homomorphically encrypt the model updates they share, an access-control step authenticates users to block internal attackers, and a server-side acknowledgment mechanism temporarily removes dropped or unresponsive users to cut waiting delay and communication. Experiments on a DR (healthcare imaging) dataset report 81.53% classification accuracy with low communication delay while preserving privacy.
Abstract¶
Healthcare data are characterized by explosive growth and value, which is the private data of patients, and its characteristics and storage environment have brought significant issues of data privacy and security. People are reluctant to share their data for privacy concerns during machine learning. To balance this contradiction, Federated Learning was proposed as a solution to train on private data without sharing it. However, many studies show that there is still the possibility of privacy leakage during the training process of federated learning. In light of this, we propose a privacy-preserving federated learning scheme with homomorphic encryption(PPFLHE). Specifically, on the client side, homomorphic encryption technology is used to encrypt the training model shared by users to ensure its security and privacy. In addition, to prevent internal attacks, Access Control (AC) technology is used to confirm the user’s identity and judge whether it is trusted; on the server side, the Acknowledgment (ACK) mechanism is designed to remove the dropped or unresponsive users temporarily, which reduces the waiting delay and communication overhead, and solves the problem of user’s exiting during training. Theoretical analysis and experimental results show that the proposed scheme achieves high data utility and classification accuracy (81.53%), and low communication delay while achieving privacy preserving, compared to state-of-the-art methods. © 2023 Elsevier B.V. All rights reserved.
Source: pdf
Extracted details¶
| Field | Value | Source | Confidence |
|---|---|---|---|
| contribution | Federated learning with homomorphically encrypted client model updates | llm | 0.60 |
| contribution | Access control to verify user identity and prevent internal attacks | llm | 0.60 |
| contribution | ACK mechanism to handle dropped or unresponsive users and reduce waiting delay | llm | 0.60 |
| schemes | Paillier | llm | 0.60 |
| datasets | DR dataset | llm | 0.60 |
| tasks | federated learning | llm | 0.60 |
| tasks | healthcare image classification | llm | 0.60 |
| threat model | federated learning with aggregation server; internal (insider) attackers and user dropout | llm | 0.60 |
| interaction | other | llm | 0.60 |
| limitations | Homomorphic encryption causes high communication overhead and ciphertext expansion | llm | 0.60 |
| limitations | Poisoning attacks on model integrity are not addressed | llm | 0.60 |
| limitations | Cannot guarantee strong consistency of model training | llm | 0.60 |
| key result | classification accuracy: 81.53% (DR dataset, federated training with HE) | llm | 0.60 |
Keyword mentions¶
Counts of vocabulary terms in the full text (first page of occurrence in brackets).
- scheme: Differential privacy 12 [p.2], Paillier 6 [p.3], CKKS 2 [p.2], Partially homomorphic encryption 2 [p.4], Leveled / somewhat HE 2 [p.3]
- model: MobileNet 9 [p.4]
- dataset: CIFAR-10 9 [p.2]
- hardware: GPU 1 [p.6]
- technique: Federated learning 48 [p.1], Taylor expansion 2 [p.5], Convolution 2 [p.6], Semi-honest model 2 [p.1]
- task: Medical / healthcare 40 [p.1], IoT / edge 4 [p.2], Finance 2 [p.9]
- topic: Homomorphic encryption 31 [p.1]
Files¶
| Path (under papers/) | Collection | Pages | Size | SHA-256 | Note |
|---|---|---|---|---|---|
references/ppml/federated-learning/wang-2023_ppflhe-privacy-preserving-federated-learning-with-he-for-healthcare-data.pdf |
ppml/federated-learning | 10 | 759.8 KB | 285b78f182a3 |
was Wang et al. - 2023 - PPFLHE A privacy-preserving federated learning scheme with homomorphic encryption for healthcare da.pdf |
Database key: ppflhe-a-privacy-preserving-federated-learning-scheme-with-homomorphic · created 2026-11-05 · updated 2026-11-05 · edit with python3 tools/paperdb.py set ppflhe-a-privacy-preserving-federated-learning-scheme-with-homomorphic COLUMN VALUE