Secure and Privacy Preserving Healthcare in the Residential Environment with Multimodal Distributed Data and Decentralized AI

SPHERE-DNA is committed to improving the security and privacy levels of distributed sensing systems used in residential environments. SPHERE-DNA leverages local differential privacy federated learning (LDP-FL) to achieve resilience against data theft and inference attacks, and also uses edge computing to avoid sensitive information transfers to remote servers. SPHERE-DNA relies on multimodal sensor technology and formulates sensor fusion as a machine learning problem that is studied together with security architectures. Hardware acceleration and approximate computing schemes will be studied to enable the deployment of the developed privacy-preservation schemes to lightweight processors and unstable networks of residential environments. SPHERE-DNA accelerates extending health condition monitoring from hospitals to homes, leading to early diagnosis and intervention, and directs healthcare towards an economically sustainable direction.