Strong evidence in macaques shows exposure route dominates tissue kinetics more than dose, so human translation depends on matching infection route and airway sampling geometry, not assuming one animal readout fits every case (DOI).
Strong tissue-mapping evidence in endomyocardial biopsies shows myocarditis can carry diffuse immune-linked injury even when white-blood-cell clusters are absent, but spot-level resolution and modest sample size limit cell-level certainty (DOI).
Large longitudinal FDNY data show a substantial first-year lung-function drop after 9/11 with an exposure-intensity gradient, but proxy exposure measures and selection differences limit causal precision (DOI).
Multi-modal evidence in ulcerative colitis-associated neoplasia links persistent bowel tissue stiffness to a cancer-primed cell state marked by YAP and MMP7, but human causality and upstream drivers remain unresolved (DOI).
A multi-country European cohort shows very high hospitalization and surgery rates in infancy and early childhood, yet incomplete prescription and inpatient diagnosis data limit mechanism and treatment inference (DOI).
In sedentary adults with obesity, one aerobic bout abolished the expected immediate postexercise rise in muscle building during amino-acid infusion, but the study was small and acute rather than clinical (DOI).
In community-dwelling adults aged 85+, malnutrition tracked lower chewing force and meal enjoyment better than the total oral score, but cross-sectional data cannot show direction of effect (DOI).
A single-center real-world cohort found lecanemab feasible with mostly asymptomatic brain-scan side effects under strict selection and MRI monitoring, but the uncontrolled design leaves cognitive benefit uncertain (DOI).
Spatial scalp maps can localize inflammatory signals near hair follicles and support niche-specific hypotheses, but they do not yet prove causal derailment for every patient (BGPT).
Entropy-based voice features modestly improved depression detection on one benchmark, but the AUC gain was small and external validation plus padding-sensitivity checks remain needed (arXiv).
In healthy adults, gamma neurofeedback reduced pain ratings and laser-evoked signals in a responder subgroup, but the effect was not clearly durable and the mechanism remains uncertain (DOI).
New scientific claims, supporting evidence, and important limitations. Every Friday. No ads.
Explore by Goal
Trending