Hristov-Kalamov, NikolaFernández-Ruiz, Raúlálvarez-Marquin, AgustínNúñez-Vidal, EstherDomínguez-Mateos, FranciscoPalacios-Alonso, Daniel2025-07-302025-07-302024-05-31Hristov-Kalamov, N., Fernández-Ruiz, R., álvarez-Marquina, A., Núñez-Vidal, E., Domínguez-Mateos, F., Palacios-Alonso, D. (2024). Comparison of an Accelerated Garble Embedding Methodology for Privacy Preserving in Biomedical Data Analytics. In: Ferrández Vicente, J.M., Val Calvo, M., Adeli, H. (eds) Artificial Intelligence for Neuroscience and Emotional Systems. IWINAC 2024. Lecture Notes in Computer Science, vol 14674. Springer, Cham. https://doi.org/10.1007/978-3-031-61140-7_28978-3-031-61139-10302-9743https://hdl.handle.net/10115/96657This research work proposes a novel, encryption-based method for comparing embeddings generated by neural networks on various information types (text, images, videos, audio, etc.). This approach prioritizes real-world applications dealing with sensitive or private data, particularly in biomedical and biometric analysis, where even minor information leaks can be highly detrimental. To address this concern, the method performs all necessary calculations within a highly secure and efficient encryption layer. Notably, this work introduces practical solutions applicable to real-world biomedical data scenarios.en-USBiomedical Data AnalyticsDeep LearningEmbedding distanceCryptographyOblivious TransferGarbled CircuitsHomomorphic EncryptionBook chapter10.1007/978-3-031-61140-7_28info:eu-repo/semantics/closedAccess