The application of AI (Artificial Intelligence) in emergency medicine helps significantly improve the quality of diagnostics under limitations of resources and time constraints in emergency cases. We have designed a comprehensive AI-based diagnostic and treatment plan decision-support system for emergency medicine by integrating the available LLMs (Large Language Models), like ChatGPT, Gemini, Claude, and others, and tuning them up with additional training on actual emergency cases. There is a special focus on early detection of life-threatening and time-sensitive diseases like sepsis, stroke, and heart attack, which are the major causes of death in emergency medicine. Additional training was conducted on a total of 600 cases (300 sepsis; 300 non-sepsis). The collective capability of the integrated LLMs is much stronger than each individual engine. Emergency cases can be predicted based on information from multiple sensors and streaming sources combining traditional IT (Information Technology) infrastructure with Internet of Things (IoT) schemes. Medical personnel compare and validate the AI models used in this work.

Background: The current manuscript’s aim was to determine the human papillomavirus (HPV) genotype-specific prevalence and distribution among individuals, males, and females, of different ages in the region of Apulia, Italy, highlighting the possible variables involved in the carcinogenicity mechanism. In addition, we proposed two hypothetical models of HPV’s molecular dynamics, intending to clarify the impact of prevention and therapeutic strategies, explicitly modeled by recent survey data. Methods: We presented clinical data from 9647 participants tested for either high-risk (HR) or low-risk (LR) HPV at the affiliated Bari Policlinic University Hospital of Bari from 2011 to 2022. HPV DNA detection was performed using nested-polymerase chain reaction (PCR) and multiplex real-time PCR assay. Statistical analysis showed significant associations for all genders and ages and both HR- and LR-HPV types. A major number of significant pairwise associations were detected for the higher-risk types and females and lower-risk types and males. Results: The overall prevalence of HPV was 50.5% (n-4.869) vs. 49.5% (n-4.778) of the study population, of which 74.4% (n-3621) were found to be HPV high-risk (HR-HPV) genotypes and 57.7% (n-2.807) low-risk HPV (LR-HPV) genotypes, of which males were 58% and females 49%; the three most prevalent HR-HPV genotypes were HPV 53 (n707-15%), 16 (n704-14%), and 31 (n589-12%), and for LR-HPV, they were 42 (19%), 6 (16%), and 54 (13%); 56% of patients screened for HPV were ≤ 30 years old, 53% were between 31 and 40 years old, 46% were 41–50 and 51–60 years old, and finally, 44% of subjects were >60 years old. Conclusions: Our study provided comprehensive epidemiological data on HPV prevalence and genotype distribution among 9647 participants, which could serve as a significant reference for clinical practice, and it implied the necessity for more effective screening methods for HPV carcinogenesis covering the use of more specific molecular investigations. Although this is a predominantly descriptive and epidemiological study, the data obtained offer not only a fairly unique trend compared to other studies of different realities and latitudes but also lead us to focus on the HPV infection within two groups of young people and adults and hypothesize the possible involvement of dysbiosis, stem cells, and the retrotransposition mechanism.

Sepsis remains a critical medical emergency caused by a dysregulated immune response to infection, with timely detection and intervention being essential for improving survival rates. Traditional methods often rely on clinician intuition and structured scoring systems, which may be time-intensive and prone to variability. To address these limitations, Machine Learning (ML) offers a powerful alternative, bringing precision and efficiency to sepsis detection. This study investigates both white-box and complex black-box ML models applied to patient data collected across the continuum of care, including monitoring at the urgent care, en route in ambulances, and diagnostics conducted within hospital emergency department settings themselves. White-box models, such as logistic regression and decision trees, are valued for their interpretability, allowing healthcare providers to understand and trust the reasoning behind predictions. Meanwhile, black-box models like deep neural networks and support vector machines deliver superior accuracy but pose challenges in clinical transparency. This trade-off between explainability and performance is explored in detail, supported by experimental results aimed at identifying the most effective computational strategies for early sepsis recognition across diverse healthcare environments.

Human microbiota, a complex consortium of microorganisms co-evolved with the host, profoundly influences tissue development, immune regulation, and disease progression. Growing evidence shows that microbial metabolites and signaling molecules modulate key stem cell pathways—such as Hedgehog, Wnt/β-catenin, and Notch—thereby reprogramming stem cell fate toward tumor-suppressive or tumor-promoting outcomes. Specific taxa within oral, intestinal, and urogenital niches have been linked to cancer initiation, therapy resistance, and recurrence. In parallel, clinical studies reveal that microbiota composition affects infection dynamics: bacterial isolates from symptomatic urinary tract infections inhibit commensal growth more strongly than the reverse, with Gram-positive and Gram-negative strains displaying distinct interaction profiles. Collectively, these findings highlight microbiota’s dual role in regulating cellular plasticity and pathogenicity. Elucidating host–microbe and microbe–microbe mechanisms may guide microbiota-targeted interventions to improve cancer and infectious disease management.