Development of Privacy-preserving Deep Learning Model with Homomorphic Encryption: A Technical Feasibility Study in Kidney CT Imaging

2025/08/27 by Sangwook Lee, Jong‐Min Choi, Min-Je Park +5 · 1 voice · 1 citation
Computer Science · #Privacy-Preserving Technologies in Data

paper · doi:10.1148/ryai.240798

openalex publication_date 2025/08/27 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/22

Abstract

Homomorphic encryption can be integrated into deep learning models for CT-based kidney mass classification, achieving diagnostic performance similar to that of nonencrypted models while ensuring privacy protection, although with increased storage and processing time demands.

Cited by

• Privacy-preserving Deep Learning in Medical Imaging: Feasibility and Challenges

Discussions

• Homomorphic encryption was integrated into deep learning models for CT-based renal mass classification https://doi.org/10.1148/ryai.240798 #DeepLearning #kidneys #HomomorphicEncryption [bsky, 5 points, 0 comments]

Related

• Application of Homomorphic Encryption in Medical Imaging
• MedBlindTuner: Towards Privacy-preserving Fine-tuning on Biomedical Images with Transformers and Fully Homomorphic Encryption
• CryptoUNets: Applying Convolutional Networks to Encrypted Data for Biomedical Image Segmentation
• Homomorphic Encryption and Federated Learning based Privacy-Preserving CNN Training: COVID-19 Detection Use-Case
• SoK: Privacy-preserving Deep Learning with Homomorphic Encryption