The study investigates the use of the Lung Ultrasound Score (LUS) as a predictive tool for assessing outcomes in patients with acute respiratory failure (ARF) caused by COVID-19, treated with non-invasive respiratory support (NIRS). Conducted in a Respiratory Intermediate Care Unit (RICU), the research highlights LUS ≥ 25 as a significant threshold for predicting 90-day mortality, failure of high-flow nasal cannula (HFNC), and the need for endotracheal intubation (ETI). Over a cohort of patients, the study found that higher LUS values were strongly correlated with poorer outcomes, offering a valuable means of early risk stratification. LUS is non-invasive, easy to calculate, and widely accessible, making it an ideal tool for clinicians to identify patients at risk of worsening conditions and optimize treatment strategies. The study emphasizes the utility of LUS in guiding clinical decision-making, particularly in resource-limited settings or intermediate care units where computed tomography (CT) may not be feasible. Additionally, the findings underscore the importance of identifying patients at risk of HFNC failure or requiring ETI to prevent delayed interventions that could lead to poorer prognosis. By proposing LUS as a critical metric in the management of COVID-19 ARF, the study advocates for its broader adoption in clinical practice. The authors also call for further multicenter studies to validate these findings and explore the application of LUS in diverse patient populations.