Unmet Need
Mechanical ventilation is used to maintain adequate gas exchange and requires clinicians to manually adjust settings based on the patient’s physiologic variables, which is a labor-intensive and time-consuming task. With the shift to automated mechanical ventilators, predominantly due to their ability to automatically change the ventilation parameters in order to maintain the patient’s physiological parameters at a certain control target, manufacturers have moved towards implementing novel algorithms, not only to distinguish themselves, but to provide new ventilation modes that can better control and manage the mechanical ventilation. However, in some marketed products, it was found that the algorithm would autonomously switch modes in response to unexpected stimuli, without providing information to the clinician, which can be fatal. Thus, although automation has the potential to decrease workload while maintaining an effective method of maintaining a mechanical ventilator’s parameters, there currently lacks a method of increasing the reliability and robustness of these algorithms to unexpected signal inputs in order to create trust in these computer algorithms.
Technology Overview
Johns Hopkins researchers have developed an autonomous mechanical ventilation system that incorporates assurance, which is the state of being clear-headed either that a hypothesis or prediction is correct or that a chosen course of action is the best or most effective, with the goal of increasing the reliability of autonomous ventilation algorithms. The system alerts a clinician when there is a potentially clinically relevant difference between the current state and target, and the addition of assuredness in the system architecture of an autonomous mechanical ventilation algorithm has the potential to significantly improve the reliability and trust in these systems for clinical use.
Stage of Development
A working prototype is under development.
Patent
N/A
Publication
