Clinical Research Directory

Development and Prospective Validation of the AutoEIT Foundation Model for ARDS

Acute Respiratory Distress Syndrome (ARDS) is a highly heterogeneous condition, making precision mechanical ventilation challenging. Electrical Impedance Tomography (EIT) provides real-time monitoring of regional lung ventilation, but interpreting its complex spatiotemporal data relies on subjective clinical experience. This multicenter ambispective cohort study aims to develop and validate an all-in-one Artificial Intelligence Foundation Model (AutoEIT). The study consists of two phases: (1) A Retrospective Phase (data from 2020 to 2025, N=1000) to train the AutoEIT model using continuous EIT images and synchronous high-frequency ventilator waveforms; and (2) A Prospective Temporal Phase (January 2026 to December 2026, N=200) to rigorously test the model's performance in a real-world clinical setting. The AutoEIT model is a deep multi-task learning framework designed to simultaneously perform five core clinical tasks: (1) classify ARDS subphenotypes, (2) recommend individualized positive end-expiratory pressure (PEEP), (3) detect occult patient-ventilator asynchrony, (4) predict ventilation/perfusion (V/Q) ratio, and (5) continuously estimate the non-invasive PaO₂/FiO₂ ratio. Ultimately, this system aims to reduce ventilator-induced lung injury (VILI) and improve precision respiratory care.

Respiratory Knowledge Portal Computer and Phone Application to Improve Quality of Mechanical Ventilation by Reducing the Number of Ventilator Associated Events, Injury Created by the Ventilator and Unsafe Setting of Alarms.

Ventilator associated events (VAE) is a quality metric defined by 48 hours of stability followed by 48 hours of escalation of ventilator settings within the ICU. VAE have been associated with poor outcomes and increases the cost of care, yet is not easy to avoid. Operationalizing all the standards of care known to improve outcomes of those requiring mechanical ventilation in the critical care environment requires a comprehensive approach. ICU teams are encouraged to follow best practice protocols to help liberate and prevent VAEs. Yet, compliance with protocols in most ICUs is suboptimal for multiple reasons. With the advent of computerized mechanical ventilators capable of streaming data from breath to breath and biomedical integration systems (BMDI) such as Capsule (UTMB's BMDI system), software systems have been developed to help identify variances in the standard of care. Automation in near real-time ventilator data feedback has been shown to reduce the incidences of VAEs. This quality improvement project will leverage Vyaire's Respiratory Knowledge Portal (RKP) to collect and store meaningful data regarding ventilator-associated events (VAE), alarm policy compliance, ventilator weaning, and lung protective analytics. Goals: 1. To collect quality metrics utilizing RKP from patients requiring mechanical ventilation over a 3-4-month period for a retrospective baseline analysis. 2. Provide the RKP tool to the ICU team to determine if the use of RKP's webportal and Messenger Zebra phone app improves quality of mechanical ventilation and outcomes. 3. To determine a return on investment (ROI) for a software system like RKP.

Biphasic Positive Airway Pressure Ventilation Versus Flow-Controlled Ventilation in Burn Patients

The goal of this clinical trial is to learn if the new ventilation mode flow-controlled ventilation (FCV) is a more protective mode of ventilation for adult patients after severe burn injury. The main question it aims to answer is: Does FCV reduce the mechanical power (a key determinant of ventilator-induced lung injury) compared to conventional pressure-controlled ventilation (PCV) during ventilation of patients with burn injury? Researchers will compare FCV with PCV for up to 70 hours of ventilation to see if the mechanical power is reduced during ventilation of participants being in need of ventilation after severe burn injury. Ventilation of participants will be controlled by either FCV or PCV. Group-specific ventilation will have the following characteristics: * FCV: Control of airway flows during inspiration and expiration, use of individualized lower and upper airway pressures and no fixed values for the volumes being inspired and expired (tidal volumes) * PCV: No control of airway flows during expiration, use of individualized lower airway pressures and upper airway pressures for a fixed tidal volume during each breath (6-8 ml/kg ideal body weight) In total, at least 24 participants in need of ventilation after severe burn injury will be ventilated either with FCV (12 participants) or PCV (12 participants) for up to 70 hours. During ventilation mechanical power is computed according to certain ventilation parameters. Additionally, we evaluate organ functions of the cardiovascular systems, the lungs and other organs during and after the group-specific ventilation.