6295 papers
This cross-sectional study of 213 recent pulmonary TB survivors in Uganda reveals that female sex, HIV co-infection, and productive cough are significant independent predictors of reduced functional exercise capacity, highlighting the need for targeted rehabilitation for these vulnerable subgroups.
This study identifies a SETDB1-linked "Dual-Vulnerability Niche" in approximately 37% of non-BRCA pancreatic ductal adenocarcinoma patients characterized by high SETDB1 expression and backup DNA repair activity, demonstrating that combined SETDB1 inhibition and olaparib treatment effectively reduces tumor growth in preclinical models and validating a machine learning classifier capable of identifying this subgroup.
This cross-sectional survey of 93 university students with ADHD indicates that a significant majority experience a transient, statistically significant improvement in self-reported attentional focus within 15 to 90 minutes following orgasm, a phenomenon that appears more pronounced in unmedicated individuals and warrants further objective investigation.
This systematic review of 56 studies (2015–2026) reveals that while AI, machine learning, and robotics demonstrate high performance in neurosurgical tasks like tumor detection and targeting, the field faces a significant translational bottleneck characterized by a high risk of bias, a lack of external validation, and a scarcity of prospective clinical trials, with robotic and adaptive systems showing greater readiness for clinical adoption than pure deep learning applications.
This cross-sectional study of 1,145 Indonesian respondents reveals that while Mpox awareness is high, willingness to pay for the vaccine is low and primarily driven by education, urban residency, prior STI history, and trust in health authorities rather than income or age, highlighting the need for targeted interventions to improve knowledge and accessibility.
This longitudinal action learning study of ten Australian GPs reveals that adopting AI scribes in residential aged care extends beyond efficiency gains to drive significant workflow redesign (the "Kaizen-to-BPR arc") and "cognitive load shedding," while simultaneously enhancing clinical communication and prompting collaborative solutions for complex consent challenges.
This case report documents the first successful use of emergency physician-performed point-of-care ultrasound (POCUS) to diagnose extensive Stanford Type A aortic dissection in a Ghanaian low-resource setting, highlighting both the diagnostic efficacy of the SPEED protocol for atypical presentations and the critical infrastructure gap preventing definitive surgical treatment.
This retrospective cohort study demonstrates that the Degree of Inflammatory Perturbation (DIP), a composite metric derived from routine laboratory parameters, serves as an independent predictor of unfavorable tuberculosis treatment outcomes, offering a low-cost tool for early risk stratification and treatment monitoring.
This scoping review synthesizes evidence from nine studies in low- and middle-income countries to demonstrate that AI chatbots and conversational agents effectively improve health knowledge and behavioral outcomes across various conditions, though their widespread implementation is hindered by digital inequities, language barriers, and a lack of rigorous clinical outcome assessments.
This systematic review reveals that while collaborative networks and capacity building facilitate One Health implementation for zoonotic disease control in Ghana, the approach is significantly hindered by severe resource constraints, weak governance, and critical policy gaps, necessitating urgent systemic investments and a comprehensive national policy framework.