Lyme disease poses an escalating global health challenge partly driven by climate change and land-use transformation, while mechanistic understanding of transmission remains limited. We analyzed prevalence records from 43 Eurasian countries to map pathogen biogeography and applied theory-driven structural equation modeling (SEM) to evaluate hypothesized direct, indirect, and cascading pathways linking climate, land cover, landscape structure, and host biodiversity to pathogen prevalence. Genospecies exhibited distinct biogeographic patterns shaped by vector-host specificity, with Ixodes ricinus dominating transmission in Europe and I. persulcatus in Asia. Borrelia afzelii prevalence was associated with climate, forest fragmentation, and landscape heterogeneity, and exhibited a negative association with rodent species richness at the continental scale. Borrelia garinii showed no host diversity effects but responded directly to temperature and landscape structure, reflecting reliance on highly mobile avian hosts. These findings reveal fundamental transmission heterogeneity among genospecies arising from divergent host-reservoir ecologies, with critical implications for disease control. Effective surveillance must integrate genospecies-specific ecology with landscape management to align biodiversity conservation, climate adaptation, and public health.
Rapid urbanization has increased overlap between human and synanthropic rodent habitats. This intensifies human-animal-environment interactions and facilitates the emergence and re-emergence of zoonotic diseases, highlighting the importance of the "One Health" concept. Calodium hepaticum (syn. Capillaria hepatica) is a neglected zoonotic parasite where rodents serve as the principal reservoirs. Human infections are often asymptomatic, but they can also cause liver failure and death in severe cases, with definitive diagnosis requiring liver biopsy. This study was conducted in Meghalaya, Northeastern India, where 173 rodents and shrews representing seven species were captured from synanthropic environments, with B. bengalensis comprising the largest proportion of captured animals (49.13%), followed by B. indica (17.34%), S. murinus (14.45%), M. musculus (8.67%), R. rattus (4.63%), M. booduga (4.05%), and R. norvegicus (1.73%). Liver samples were examined using gross pathology, histopathology, parasitological, and molecular methods, including 18S rRNA gene sequencing and phylogenetic analysis. Histopathological examination revealed non-embryonated barrelshaped eggs in hepatic tissues, while molecular analysis confirmed the identity of C. hepaticum. Based on molecular detection, the overall prevalence of C. hepaticum was 36.99% (64/173; 95% CI: 30.16-44.44). Specieslevel prevalence was highest in Bandicota indica (80%; 95% CI: 62.69-90.49), followed by Mus booduga (57.14%; 95% CI: 25.05-84.18), B. bengalensis (31.76%; 95% CI: 22.84-42.27), and M. musculus (26.67%; 95% CI: 10.90-51.95), with lower rates in S. murinus (16%; 95% CI: 6.40-34.65) and R. rattus (12.50%; 95% CI: 2.24-47.09). Given the high prevalence observed, appropriate rodent control and carcass disposal should be encouraged to minimize the potential risk of environmental contamination, as eggs may be released during decomposition, predation, or cannibalism and subsequently embryonate in soil, becoming infective to humans or other susceptible hosts.
Antimicrobial resistance (AMR) poses a critical global health challenge, leading to substantial mortality and economic losses without science-based interventions. Conventional approaches to combating AMR have important limitations in addressing this escalating threat. Recent advances in AI, including deep learning, large language models, and protein language models, provide new opportunities to integrate multimodal data for AMR research. However, evidence remains uneven, with many approaches limited to computational benchmarking rather than real-world validation. Here, we critically review AI applications in AMR, focusing on rapid diagnosis, therapeutic discovery, and clinical decision support. AI-based diagnostic models can identify resistant pathogens and predict resistance profiles, but most lack prospective clinical evaluation. AI-assisted drug discovery and alternative therapies have generated promising candidates, although most remain at computational, in vitro, or early preclinical stages. Large language models may support decision-making, but their clinical use remains preliminary. Successful translation of AI into clinical and One Health practice will require representative datasets, independent validation, prospective evaluation, and robust governance.
Background Tetanus remains a vaccine-preventable but potentially fatal disease, with control relying on vaccination coverage, maternal/neonatal protection, adult boosters, wound prophylaxis, and clinical management. This study characterized the global research landscape, thematic evolution, and emerging frontiers of tetanus literature from 2005 to 2025. Methods English-language articles and reviews were retrieved from the Web of Science Core Collection (WoSCC) and Scopus, with the final search conducted on May 12, 2026. Records were merged, deduplicated, and manually screened. Bibliometrix, VOSviewer, CiteSpace, and Scimago Graphica were used for descriptive analyses and visualization of collaboration, keyword co-occurrence, co-citation, and burst patterns. Citation-trend and reference co-citation analyses were restricted to WoSCC records. A manually verified clinical-treatment subset underwent supplementary keyword analysis. Results Of 28,912 retrieved records, 8307 publications met the inclusion criteria, comprising 7689 articles and 618 reviews. Annual publication output remained relatively stable, with a modest increase after 2010. The United States contributed the most publications (2578; 31.03%), followed by the United Kingdom, India, Germany, and China. The Centers for Disease Control and Prevention was the most productive institution, and Vaccine was the leading journal (643 publications). Three broad thematic domains emerged: toxin biology and immune mechanisms, vaccines and immunization strategies, and public health implementation and population protection. Recent keyword bursts highlighted vaccination coverage, pregnancy, women, vaccine hesitancy, and public health, while mechanistic and immunological research remained active. The clinical-treatment subset included 675 publications and emphasized intensive care, antitoxin and immunoglobulin therapy, autonomic dysfunction, and mortality. Conclusions Recent tetanus research fronts increasingly emphasize immunization delivery and population protection, alongside continuing mechanistic research. Clinical management constitutes a distinct but less prominent subdomain. These findings provide a publication-level overview to inform future evidence synthesis and research planning but should be interpreted alongside clinical, epidemiological, and surveillance evidence.
