Healthy aging induces parallel changes in brain functional activity and structural morphology, yet the interplay between these changes remains unclear. Prof. Yuhui Du’s team at the College of Computer and Information Technology, Shanxi University, in collaboration with Prof. Vince D. Calhoun (Georgia State University), analyzed multimodal neuroimaging data from 27,793 healthy subjects (aged 49-76 years) in the UK Biobank. They proposed a unified framework for single-modal and multimodal brain-age prediction and joint functional-structural aging analysis, systematically characterizing diverse synergistic vs. contradictory aging patterns between functional network connectivity (FNC) and gray matter volume (GMV). Importantly, these joint patterns were further linked to specific cognitive decline. The study, titled “Joint aging patterns in brain function and structure revealed using 27,793 samples” was published in Research (2025, 8:0887; DOI: 10.34133/research.0887).

Abstract

Purpose – The impact of digital transformation on banks’ systemic risk merits thorough investigation.

Design/methodology/approach – This study examines the influence of digital transformation on banks’ systemic risk based on the fixed effect model with quarterly unbalanced panel data on 36 listed commercial banks in China from 2011 to 2020.

Findings – Results show that digital transformation has a negative impact on banks’ systemic risk by reducing both bank-specific tail risk and systemic linkage to extreme market shocks. Heterogeneity analysis suggests that digital transformation can significantly reduce systemic risk in national commercial banks relative to regional commercial banks, mediated through lowered management costs. Finally, this study finds an asymmetric relationship between digital transformation and banks’ systemic risk. Particularly, a desirable level of digital transformation can reduce systemic risk, while excessive digital transformation may exacerbate it.

Originality/value – These findings provide valuable guidance for promoting digital transformation for banks and mitigating systemic risk from digitalization.

Thermoelectric technology that utilizes thermodynamic effects to convert thermal energy into electrical energy has greatly expanded wearable health monitoring, personalized detecting, and communicating applications. Encouragingly, thermoelectric technology assisted by artificial intelligence exerts great development potential in wearable electronic devices that rely on the self-sustainable operation of human body heat. Ionic thermoelectric (i-TE) devices that possess high Seebeck coefficients and a constant and stable electrical output are expected to achieve an effective conversation of thermal energy harvesting. Herein, we developed an i-TE paster for thermal chargeable energy storage, temperature-triggered material recognition, contact/non-contact temperature detection, and photo thermoelectric conversion applications. An all-solid-state organic ionic gel electrolyte (PVDF-HFP-PEO gel) with onion epidermal cells-like structure was sandwiched between two electrodes, which take full advantage of a synergy between the Soret effect and the polymer thermal expansion effect, thus achieving the enhanced ZT value up to 900% compared with the PEO-free electrolyte. The i-TE device delivers a Seebeck coefficient of 28 mV K⁻¹, a maximum energy conversion efficiency of 1.3% in performance, and ultra-thin and skin-attachable properties in wearability, which demonstrate the great potential and application prospect of the i-TE paster in self-sustainable wearable electronics.

Professor Chuan He's research group at Southern University of Science and Technology reported an example of asymmetric Si–H/O–H coupling between racemic monohydrosilanes and alcohols in the same catalytic system, simultaneously achieving enantiomeric construction of the silicon chiral center and precise control of the Z/E configuration of the alkene. Through mechanistic studies combined with DFT calculations, the stereopolymerization of silicon chirality and the cis-trans isomerization process of the alkene were elucidated in detail. This reaction exhibits excellent yields and good to excellent enantiomeric selectivity, providing a new scheme for the efficient synthesis of four stereoisomers [( R,Z), (R,E), (S,Z), (S,E)]. The article was published as an open access Research Article in CCS Chemistry, the flagship journal of the Chinese Chemical Society.

Graphs are widely used to represent complex relationships in everyday applications such as social networks, bioinformatics, and recommendation systems, where they model how people or things (nodes) are connected through interactions (edges). Subgraph matching—the task of finding a smaller pattern, or query subgraph, within a larger graph—is crucial for detecting fraud, recognizing patterns, and performing semantic searches. However, current research on streaming subgraph, a similar task where timing is important, matching faces major challenges in scalability and latency, including difficulties in handling large graphs, low cache efficiency, limited query result reuse, and slow indexing performance. To address these issues, Liuyi Chen et al. presented a new framework that leverages a subgraph index based on graph embeddings, enabling effective caching and reuse of query results while demonstrating robustness and consistency across varying batch sizes and datasets. Their work was published in Intelligent Computing, a Science Partner Journal, under the title “Accelerating Streaming Subgraph Matching via Vector Databases”.