Daily Respiratory Research Analysis

In a multicenter diagnostic validation (n=176), trained nonexpert healthcare professionals using AI guidance achieved diagnostic-quality 8-zone lung ultrasound in 98.3% (95% CI 95.1%-99.4%) of cases, statistically indistinguishable from expert-acquired studies (difference 1.7%; 95% CI -1.6% to 5.0%). Masked expert readers validated image quality, and analyses followed intention to treat.

Impact: Demonstrates that AI can standardize and democratize LUS acquisition, potentially expanding access to high-value cardiopulmonary imaging in resource-limited settings.

Clinical Implications: Enables task-shifting of LUS acquisition to trained nonexperts, supporting triage for dyspnea, heart failure, and pulmonary edema. Could underpin scalable point-of-care ultrasound programs and tele-expertise workflows.

Future Directions: Evaluate impact on clinical decision-making, outcomes, and cost-effectiveness; assess performance in diverse settings, including prehospital and low-resource environments; integrate with tele-ultrasound workflows.

Using >200 isogenic barcoded SARS-CoV-2 variants in hamsters, longer and more direct exposures increased the transmission bottleneck, and barcode enumeration across upper and lower airways showed the upper airway as the primary source of transmitted virus. The barcoded-virus platform provides a rigorous tool to quantify transmission dynamics.

Impact: Introduces a powerful methodological platform to dissect transmission events and quantify bottlenecks, informing vaccine and nonpharmaceutical interventions and viral evolution models.

Clinical Implications: While preclinical, findings support targeting upper airway viral load and minimizing prolonged/direct exposure in infection control. Barcoding approaches can inform variant fitness and transmission risk assessments.

Future Directions: Translate to human challenge or household transmission studies; integrate with immune status and variant-specific fitness; apply to other respiratory viruses to generalize bottleneck principles.

Among 21,062 never-smokers undergoing opportunistic LDCT in South Korea, screen-detected lung cancers were predominantly stage 0–I adenocarcinoma in both sexes, with no significant sex-based differences in diagnosis rates, stage, histology, or lung cancer–specific death. Findings suggest similar overdiagnosis risks for men and women and limited benefit from indiscriminate LDCT in never-smokers.

Impact: Offers high-quality, sex-stratified evidence on LDCT screening in never-smokers, informing policy debates in regions with rising INS incidence and widespread opportunistic screening.

Clinical Implications: Supports reframing LDCT screening policy for never-smokers toward targeted risk-based strategies rather than indiscriminate screening, to mitigate overdiagnosis.

Future Directions: Develop and validate risk models for never-smokers to guide targeted LDCT; evaluate net benefit including overdiagnosis and radiation harms; assess environmental and genetic risk modifiers.