Single-cell analysis of human PBMCs in healthy and type 2 diabetes populations: dysregulated immune networks in type 2 diabetes unveiled through single-cell profiling.

Abnormalities in glucose metabolism that precede the onset of type 2 diabetes (T2D) activate immune cells, leading to elevated inflammatory factors and chronic inflammation. However, no single-cell RNA sequencing (scRNA-seq) studies have characterized the properties and networks of individual immune cells in T2D. Here, we analyzed peripheral blood mononuclear cells (PBMCs) from non-diabetes and T2D patients by scRNA-seq. We found that CD14 monocytes in T2D patients were in a pro-inflammatory state and intermediate monocytes expressed more MHC class II genes. In T2D patients, cytotoxic CD4 T cells, effector memory CD8 T cells, and γδ T cells have increased cytotoxicity and clonal expansion. B cells were characterized by increased differentiation into intermediate B cells, plasma cells, and isotype class switching with increased expression of soluble antibody genes. These results suggest that monocytes, T cells, and B cells could interact to induce chronic inflammation in T2D patients with pro-inflammatory characteristics.

Human-Unrecognizable Differential Private Noised Image Generation Method.

Differential privacy has emerged as a practical technique for privacy-preserving deep learning. However, recent studies on privacy attacks have demonstrated vulnerabilities in the existing differential privacy implementations for deep models. While encryption-based methods offer robust security, their computational overheads are often prohibitive. To address these challenges, we propose a novel differential privacy-based image generation method. Our approach employs two distinct noise types: one makes the image unrecognizable to humans, preserving privacy during transmission, while the other maintains features essential for machine learning analysis. This allows the deep learning service to provide accurate results, without compromising data privacy. We demonstrate the feasibility of our method on the CIFAR100 dataset, which offers a realistic complexity for evaluation.