Thursday, April 3, 2025
4/3/2025
Novel physiomarkers of high-risk labored breathing for advance warning of clinical deterioration - Respiratory failure is common (800,000 yearly intubations), life-threatening (35% mortality rate), and costly (12% of all hospital costs). Advance warning of respiratory deterioration may trigger timely treatment, but it is difficult to predict an imminent deterioration despite advanced datasets and analytics. To address this problem, the proposed project will address a critical gap in contemporary physiological monitoring: the inability to quantify labored breathing. Breathing patterns (or respiratory kinematics) contain vital prognostic information that is not captured by conventional monitoring. There has been no way to quantify this information in a clinical practice setting. As such, this crucial dimension of pathophysiology has neither been rigorously researched nor has it been integrated into any early warning score. Most insights about the significance of respiratory kinematics are based on experts’ opinions and on small physiological studies. Little is known about the real- world distribution of any respiratory kinematic abnormality or its sensitivity, specificity, or predictive value for respiratory failure. The barriers to large-scale respiratory kinematic measurement were recently overcome with the development of a simple and scalable new method called ARK (Analysis of Respiratory Kinematics). ARK uses wearable motion sensors that are easy to apply in any setting. Yet, powered by original breakthroughs in inertial signal processing, it quantifies respiratory kinematics with high fidelity. Building on this groundbreaking innovation, the proposed study will pioneer the quantitative documentation of the human respiratory kinematic profile. The kinematic properties of the mature respiratory system will be recorded in a large cohort of adults during emergency room visits for respiratory illness. Features of the kinematic signals will be selected using both clinically-informed and data-driven approaches. In the clinically-informed approach, the feature selection will be based on an explainable correspondence with motion patterns that are established to be pathognomonic of high-risk labored breathing. In data-driven feature selection, advanced analytic toolkits will be used for large-scale, automated feature extraction, and the most informative features will be selected without regard to visual correlates. High-dimensional kinematic phenotypes will be elucidated by analyzing variability patterns across multiple features. Finally, this project will achieve the first-ever validation of ARK in a non-adult population by establishing the safety, feasibility, and validity of ARK monitoring in a cohort of premature neonates. The main hypothesis is that novel respiratory kinematic physiomarkers (features and phenotypes) will improve the prediction of severe respiratory failure beyond conventional monitoring. The proposed work is expected to result in a comprehensive understanding of the diagnostic and prognostic significance of well-known breathing motion patterns, and the discovery of novel breathing phenotypes. This work holds the promise for a far-reaching impact, including major leaps in predictive analytics that trigger timely treatment, reduce ventilator use, unburden ICUs, and save lives.
HHS’ Tracking Accountability in Government Grants System (TAGGS) website provides access to detailed descriptions of grants, loans, aggregated direct payments and other types of financial assistance awarded by the HHS Operating Divisions (OPDIVs) and the Office of the Secretary Staff Divisions (STAFFDIVs).