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

. Purpose To evaluate the technical feasibility of implementing homomorphic encryption in deep learning models for privacy-preserving CT image analysis of renal masses. Materials and Methods A privacy-preserving deep learning system was developed through three sequential technical phases: a reference CNN model (Ref-CNN) based on ResNet architecture, modification for encryption compatibility (Approx-CNN) by replacing ReLU with polynomial approximation and max-pooling with averagepooling, and implementation of fully homomorphic encryption (HE-CNN).

Rechargeable Aluminum-Ion Batteries Based on an Open-Tunnel Framework.

Rechargeable batteries based on an abundant metal such as aluminum with a three-electron transfer per atom are promising for large-scale electrochemical energy storage. Aluminum can be handled in air, thus offering superior safety, easy fabrication, and low cost. However, the development of Al-ion batteries has been challenging due to the difficulties in identifying suitable cathode materials.