The iSAVE has been validated using benchtop models as well as in vivo studies on large mammals.

The results of our validation have been published in Science Translational Medicine and can be found here.

This page provides a brief overview of the validation. 


Benchtop Validation on ICU (closed-circuit) Ventilators

Hamilton G5 Puritan Bennet 840 Philips VX850 The iSAVE was connected to several common models of ICU ventilators to ensure a generalizable circuit and utilized to ventilate two test lungs.


Differential Tidal Volume Delivery

By titrating the flow control valve, the iSAVE can achieve differential tidal volumes (VT) spanning ratios from 50:50 to 15:85. 


Differential PEEP

By titrating the PEEP valves, the iSAVE can achieve differential PEEP. Here, we ventilated the two lungs at a PEEP of 5 cm H2O (blue lung) and 10 cm H2O (black lung). 


In Vivo Validation: Ventilation of Two Large Animals on iSAVE

We investigated whether the iSAVE could ventilate two large animals as adequately as ventilating each animal on its own ventilator. For a period of 45 minutes, animals were ventilated stably on the iSAVE. Their oxygenation and status were not different from those when they were individually ventilated. 


iSAVE Enables Rebalancing of Ventilation in Response to Changes Induced by a Patient's Improvement or Deterioration

In response to:
B) Decreased compliance
C) Increased compliance
D) Increased resistance
E) Decreased resistance
occuring in the regions shaded red. Titration of the flow valves enables return of baseline ventilation parameters (green shaded region).


Valves and Filters Prevent Cross Contamination

Given the high aerosolization risk of the coronavirus, it is critical that the iSAVE prevent cross contamination under potentially turbulent or unusual airflow patterns. We simulated a contaminant entering the circuit using a nebulizer (A). Even at unrealistic conditions yielding turbulent flows at high pressures (nebulization at 40cmH2O, 10 minutes of continuous nebulization, 5 mL of nebulized particles, VT = 300 mL, RR = 30 bpm), no cross contamination of filters was visualized observed (B) or detected in wipe tests of each segment of the circuit (C). 



Our benchtop and in vivo results demonstrate the potential for iSAVE to deliver multiplexed and personalized ventilation through a system of valves and sensors, with embedded safety measures. Our benchtop and in vivo testing simulated the characteristics of patients with ARDS as well as clinical scenarios associated with multiplexing ventilation. The results show that the system enables patient-specific titration of VT, inspiratory pressure, and PEEP, significantly amplifying the capacity of a single ventilator. In a set of scenarios mimicking the clinical evolution of patients with ARDS and sudden events that could jeopardize the ventilation of other patients, the system is able to handle circuit dependencies and deliver the desired therapeutic parameters to each channel. These results indicate the potential for rapid translation into clinical use to address the current ventilator shortage as a result of the COVID-19 pandemic.