Reperfusion therapy is the mainstay of treatment for ischemic stroke (IS) but frequently exacerbates secondary injury. Following cerebral ischemia and hypoxia, lactate accumulates markedly. Traditionally regarded as a metabolic byproduct, lactate has gained new significance with the discovery of protein lactylation. In addition to experimental evidence, recent computational models have been developed to predict potential proteins and sites of lactylation. Nevertheless, the roles of lactate and lactylation in cerebral ischemia reperfusion (I/R) injury remain unclear and even controversial. Current studies suggest that ischemic and reperfusion phases differ in their pathological changes, and accumulating evidence indicates that non-histone lactylation may aggravate injury. Proposed mechanisms include the promotion of neuronal death, exacerbation of neuroinflammation, modulation of mitochondrial function, and alteration of angiogenesis. Furthermore, potential crosstalk between lactylation and other post-translational modifications, whether synergistic or antagonistic, may influence disease progression, yet such interactions in cerebral I/R injury have not been systematically investigated. Current strategies to modulate lactylation include targeting lactate levels, lactate dehydrogenase, transporters, and modifying enzymes. This review summarizes multiple factors influencing non-histone lactylation and discusses potential intervention strategies, highlighting their advantages and limitations and providing new perspectives for the treatment of cerebral I/R injury.

A novel hybrid meta-atom structure, utilizing atomic layer-coated nanoparticle-embedded resin (nano-PER), has been introduced to enable the scalable manufacturing of high-performance metasurfaces. By using TiO₂ nano-PER as a core material and coating it with a thin TiO₂ layer, researchers successfully implemented the optical properties of high-index materials with a printable composite. This approach significantly reduced the meta-atom height by over 27% compared to previous hybrid resin-based structures, mitigating risks like structural bending and collapse during fabrication. This breakthrough, demonstrated through the fabrication of a hyperbolic metalens via Nanoimprint Lithography (NIL) and Atomic Layer Deposition (ALD), paves the way for the practical, high-throughput integration of metasurfaces into various optical applications.

A team of researchers from the Federal University of Technology Owerri in Nigeria has highlighted a promising new way to cut diesel engine pollution without sacrificing performance. Their review of global research on Water-in-Diesel Emulsion (WiDE) technology shows that blending small amounts of water into diesel fuel can dramatically reduce harmful emissions while maintaining or even improving engine efficiency.

Researchers from the University of Toronto have uncovered how the hidden architecture of wood can give rise to biochar materials as strong as mild steel. Their new study reveals that the direction in which biochar is measured can make its hardness vary by more than twenty-eight times, opening the door to a new generation of sustainable carbon materials for use in energy devices, filters, and structural applications.

A sweeping new study has uncovered global patterns in how bacteria thrive and interact within lakes and reservoirs, offering new insights into the invisible forces that sustain freshwater ecosystems. The research, led by scientists from Xi’an University of Architecture and Technology, analyzed hundreds of samples from around the world to understand how geography, temperature, and nutrients shape bacterial communities in water and sediments.

Professor Lijia Wang's group at East China Normal University, in collaboration with Academician Yong Tang of the Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences, developed a nickel (II)-catalyzed asymmetric [2+2] cyclobutanization reaction. By introducing a flexible "antenna" structure into the traditional chiral BOX ligand, they successfully achieved efficient asymmetric cyclobutanization of indole-derived heterocyclic enamines with methylene malonates. Based on this method, the team completed a concise asymmetric synthesis of seven quebracho indole alkaloids with a total yield of 29%-39% in just 7-10 steps (the longest linear step). Among them, (-)-eburine, (-)-eburcine, and (-)-minovine were achieved for the first time in asymmetric total synthesis, providing a new path for the efficient preparation of complex alkaloids. The article was published as an open access research article in CCS Chemistry, the flagship journal of the Chinese Chemical Society.

Nowadays, compute-intensive programs, like those for training artificial intelligence and machine learning models, are used extensively. Modern compilers use vectorization techniques to exploit parallel processing capabilities to improve the performance of such programs. A group of scientists from the University of Southern California, Cisco AI Research, and Intel Labs designed a data-driven, graph-based learning framework for automatic vectorization called autograph, which utilizes deep reinforcement learning to have an intelligent agent learn an optimal policy. Autograph greatly outperformed other approaches across different datasets. The work was published on June 2, 2025, in an article titled “A Graph-Based Learning Framework for Compiler Loop Auto-Vectorization” in Intelligent Computing, a Science Partner Journal.

Critical concerns regarding the security and privacy of information transmitted within Internet of Medical Things systems have increased greatly, since these systems manage and generate substantial amounts of sensitive private data. Current traditional security methods have not yet adapted to evolving cyber threats, making the need for data security in medical settings crucial. Recently, a security framework based on blockchain technology and distributed reinforcement learning has been presented to address these challenges. The new framework ensures that data are stored securely and transmitted reliably while minimizing resource usage, and it also enables security measures to adapt to changing threat patterns and enhance system resilience against attacks. In summary, the new methods demonstrated improved memory consumption and transaction latency compared to existing approaches, while maintaining high data throughput. This work was published in Intelligent Computing, a Science Partner Journal, under the title “Privacy-Preserving Strategies in the Internet of Medical Things Using Reinforcement Learning and Blockchain” by Dounia Doha and Ping Guo.

A research paper by scientists at The Chinese University of Hong Kong presented a new robotic bimodal palpation technique utilizing circumferential and sliding B-scan modes along with Bayesian optimization for precise lesion center and boundary detection..

The research paper, published on Sept. 2, 2025 in the journal Cyborg and Bionic Systems.

A research paper by scientists at Southern Medical University presented a novel federated metadata-constrained method with mutual learning for all-in-one CT reconstruction.

The research paper, published on Aug. 27, 2025 in the journal Cyborg and Bionic Systems.