A deep learning system (DL-T1b) leveraging digital pathology achieved strong performance (AUC 0.910) for predicting lymph node metastasis in T1b gastric cancer. Integration with six clinical risk factors into a nomogram further boosted predictive accuracy, offering a precise tool for personalized treatment decisions. This study establishes proof-of-concept for deep learning-based pathomics in early gastric cancer risk stratification.

A new Editorial published in Translational Exercise Biomedicine (ISSN: 2942-6812), an official partner journal of International Federation of Sports Medicine (FIMS), argues that physical inactivity can no longer be described merely as a public-health crisis. Instead, the authors contend, it represents a persistent implementation failure of modern societies to align with fundamental human biology with consequences that strain global health systems, economic sustainability, and long-term human resilience. The editorial was authored by Yannis P. Pitsiladis, Daria Obratov, Fabio Pigozzi and Ugur Erdener.

Sight-threatening antibiotic-resistant eye infections are becoming a significant threat to vision in dogs, cats, and horses, according to a newly published comprehensive review. The study examines global trends in antimicrobial resistance in animal eye infections and warns that multidrug-resistant bacteria are becoming more common in referral veterinary settings. The review also highlights a critical challenge for veterinarians: standard laboratory tests may not always predict how well topical eye treatments will work in practice. The authors call for earlier diagnostic testing, more targeted antibiotic use, stronger infection-control measures, and the development of antibiotic-sparing strategies to help preserve treatment options for future patients.

A joint research team from China Pharmaceutical University, Chinese Academy of Agricultural Sciences, National Engineering Research Center for the Emergency Drug, and Shandong Agricultural University has developed a novel multisite bridging‑mediated lateral flow immunoassay (mbLFIA) that enables ultrasensitive, enzyme‑free, visual on‑site detection of mosquito‑borne viruses such as chikungunya virus (CHIKV). The breakthrough, published in Targetome, addresses a critical global need for field‑deployable diagnostics in resource‑limited regions, where conventional nucleic acid tests rely on bioenzymes and bulky laboratory instruments.