Beijing Haidian Hospital

Abstract Tungsten has demonstrated a competitive figure of merit in its application to plasma-facing components (PFCs) of fusion reactors. During service, the material is exposed to high temperatures and high-level displacement damage. A common interest is fostered in the nuclear materials community to address the issue of defect evolution at operating temperatures, and how they recover throughout service. During maintenance, the application of in situ thermal repair technologies is tempting, featuring attractive efficiency in defect removal via an optimal selection of post-irradiation annealing (PIA) parameters. In previous studies, we examined the role of PIA temperature, PIA duration, and initial defect concentration on defect evolution, and redefined the damage recovery stages for tungsten, but this was done from a room-temperature heavy-ion irradiation perspective; see Wang et al (2023 J. Nucl. Mater. 581 154454), Wang et al (2024 Acta Mater. 273 119942). In this study, the scope is expanded to displacement damage saturation induced by heavy-ions at high temperatures, relevant to the service conditions of tungsten-based PFCs. The damage microstructure evolution in response to varied irradiation temperatures ( T Irr ) and PIA temperatures ( T PIA ) was assessed via transmission electron microscopy and Doppler broadening positron annihilation spectroscopy. Irradiation hardening was evaluated via nano-indentation. A scientific framework is proposed to guide thermal healing of displacement damage in tungsten via PIA treatment. It was ineffective when T PIA ⩽ T Irr . An adverse effect of PIA-induced secondary hardening occurred when T PIA (stage III) > T Irr (stage III). The optimal PIA scheme was confirmed when T PIA (stage IV) > T Irr (stages III–IV), eluding PIA-induced secondary hardening and minimizing PIA-enhanced recrystallization.

While large language models (LLMs) leverage both knowledge and reasoning during inference, the capacity to distinguish between them plays a pivotal role in model analysis, interpretability, and development. Inspired by dual-system cognitive theory, we propose a cognition attribution framework to decouple the contribution of knowledge and reasoning. In particular, the cognition of LLMs is decomposed into two distinct yet complementary phases: knowledge retrieval (Phase 1) and reasoning adjustment (Phase 2). To separate these phases, LLMs are prompted to generate answers under two different cognitive modes, fast thinking and slow thinking, respectively. The performance under different cognitive modes is analyzed to quantify the contribution of knowledge and reasoning. This architecture is employed to 15 LLMs across 3 datasets. Results reveal: (1) reasoning adjustment is domain-specific, benefiting reasoning-intensive domains (e.g., mathematics, physics, and chemistry) and potentially imparing knowledge-intensive domains. (2) Parameter scaling improves both knowledge and reasoning, with knowledge improvements being more pronounced. Additionally, parameter scaling make LLMs reasoning significantly more prudent, while moderately more intelligent. (3) Knowledge primarily resides in lower network layers, while reasoning operates in higher layers. Our framework not only helps understand LLMs from a "decoupling" perspective, but also provides new insights into existing research, including scaling laws, hierarchical knowledge editing, and limitations of small-scale-LLM reasoning.

BackgroundTetanus is a severe but vaccine-preventable disease controlled by tetanus toxoid (TT) immunization and tetanus immunoglobulin (TIG) therapy. Although maternal–neonatal tetanus has been eliminated in many regions, coverage gaps and sporadic cases persist in low-resource settings. As a longstanding component of global immunization and a key conjugate carrier platform, tetanus research remains essential despite advances in novel vaccine technologies. However, the global research landscape on tetanus vaccines and immunoglobulins remains unmapped.ObjectiveThis study aimed to conduct a comprehensive bibliometric analysis of publications on TT and TIG from 2000 to 2025, to characterize research dynamics, leading contributors, and thematic evolution.MethodsWe retrieved publications from the Web of Science Core Collection (SCI-Expanded) for 2000–2025 using relevant keywords (“tetanus vaccine*” OR “tetanus toxoid*” OR “tetanus immunoglobulin*” OR “tetanus immune globulin*” OR “tetanus antitoxin*”). Following screening, 2,949 English-language articles and reviews were included. Bibliometric analyses (publication trends, funding agencies, co-authorship, co-citation, and keyword co-occurrence) were performed using VOSviewer, CiteSpace, and the bibliometrix R package.ResultsA total of 2,949 publications were analyzed. Annual output remained stable (mean ≈ 113.4), with citations peaking in 2021 (n=7,291). The USA led in volume (1,130 papers), while Switzerland achieved the highest impact (88.41 citations/article). The Centers for Disease Control and Prevention (CDC) and University of Oxford emerged as central hubs, with GlaxoSmithKline (GSK) illustrating the role of industry-academic partnerships. A strategic funding shift occurred around 2010: while the National Institutes of Health (NIH) drives foundational research, the Bill & Melinda Gates Foundation (BMGF) has become the engine for translational equity. Evolutionary trends show a shift from early cellular mechanisms toward maternal immunization coverage, implementation determinants in low- and middle-income countries (LMICs), and the expanding utility of tetanus toxoid (TT) as a conjugate vaccine carrier.ConclusionTetanus research has evolved into a dynamic model for modern vaccinology, bridging classical immunization and next-generation technology. Beyond maternal protection, the field is expanding toward conjugate platforms and recombinant biologics. Achieving global elimination requires aligning these biotechnological innovations with equitable implementation strategies in low-resource settings.

Diabetic retinopathy (DR) is a prevalent chronic complication of type 2 diabetes mellitus (T2DM), contributing significantly to vision impairment. Circular RNAs (circRNAs) have emerged as key regulators in the pathogenesis of DR. This study aimed to investigate the differential expression profiles and potential functions of circRNAs in patients with DR compared to those without DR. In this single-center, prospective study, serum samples from 20 T2DM patients with DR and 20 without DR were analyzed using the Arraystar Human circRNA Expression Profile V2.0 microarray. An additional cohort of 40 patients (20 DR and 20 without DR) was used to validate selected circRNAs via quantitative reverse transcription PCR (qRT-PCR). Enriched signaling pathways were identified using Gene Ontology and Kyoto Encyclopedia of Genes and Genomes analyses. CircRNA-miRNA-mRNA interaction networks were constructed using bioinformatics tools. A total of 376 circRNAs were differentially expressed, including 169 upregulated and 207 downregulated circRNAs. qRT-PCR validation confirmed five downregulated circRNAs and three upregulated circRNAs, which were selected for competing endogenous RNA analysis. The circRNA-miRNA-mRNA network analysis revealed that hsa_circRNA_407262 (also known as circRNA LOC389906) may act as a miRNA sponge regulating key signaling pathways, including wingless/integrated, gonadotropin-releasing hormone, tumor necrosis factor, and rat sarcoma pathways. This study identified five downregulated and three upregulated circRNAs associated with DR. Among them, hsa_circRNA_407262 may play a crucial role in the pathological process of DR and holds potential as a candidate biomarker for early diagnosis.