Privacy-Preserving Federated Learning for Medical Image Classification with Selective Homomorphic Encryption
AI · 15.07.2026 · 7 (7) · 264
DOI: 10.3390/ai7070264 ↗Kurz gefasst
Fachartikel in der Zeitschrift AI, online erschienen am 15.07.2026. Verfasst von Zhaobin Li und 3 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Beijing Natural Science Foundation, Fundamental Research Funds for the Central Universities.
Abstract (Originalsprache)
Federated learning lets hospitals train shared diagnostic models without exchanging patient images, yet the updates they exchange each round can be inverted to reconstruct training images. Homomorphic encryption (HE) protects these updates, but encrypting an entire model with CKKS inflates communication and computation to impractical levels for cross-silo medical use. We present PASHE-FL, which exploits the structure of personalized federated learning: the client-specific classifier head stays local and is never uploaded, so encryption need only cover the shared backbone. The server ranks backbone coordinates by importance from the public global model and selects the same top-ρ set for all clients, avoiding mask negotiation; these are encrypted with CKKS, the remaining coordinates are quantized, and the encrypted fraction is annealed over training. On four medical image-classification tasks, PASHE-FL matches the personalized FedPer baseline within about one accuracy point while cutting per-round uplink by roughly 7.1–7.6× and encryption time by about 8× relative to full-model HE; this accuracy comes from personalization, not encryption. Under a gradient-inversion attack, encrypting only the top 5–10% most important coordinates collapses reconstruction quality, whereas encrypting random coordinates does not. PASHE-FL offers an empirical privacy–cost trade-off under the stated threat model rather than a formal privacy guarantee.
Förderung
- Beijing Natural Science Foundation
- Fundamental Research Funds for the Central Universities
Zitieren
APA
Li, Z., Mo, M., Du, C., & Wei, Z. (2026). Privacy-Preserving Federated Learning for Medical Image Classification with Selective Homomorphic Encryption. AI, 7(7), 264. https://doi.org/10.3390/ai7070264
Vancouver
Li Z, Mo M, Du C, Wei Z. Privacy-Preserving Federated Learning for Medical Image Classification with Selective Homomorphic Encryption. AI. 2026;7(7):264. doi:10.3390/ai7070264
BibTeX
@article{Li2026,
author = {Li, Zhaobin and Mo, Mingliang and Du, Chenchong and Wei, Zhanzhen},
title = {Privacy-Preserving Federated Learning for Medical Image Classification with Selective Homomorphic Encryption},
journal = {AI},
year = {2026},
volume = {7},
number = {7},
pages = {264},
doi = {10.3390/ai7070264}
}
EUVIMED