Limitations & Future Work

Overview

VentCU ultimately aims to create a realistic, clinically viable option for medical professionals working in the field. VentCU does not yet meet this goal and requires further development and testing. We encourage others to take the concepts and learnings from this design and take them further, to a standard that will be considered acceptable for clinical use.

Motor Selection Re-evaluation

The NEMA 23 stepper motor currently used is under-torqued, meaning the device currently cannot reach the desired breaths per minute and inspiration to expiration ratio targets. At its recommended specifications, the device should be able to reach 30 breaths per minute at an I:E ratio of 1:1, 1:2, 1:3, and 1:4. Currently, we are only able to reach BPMs of 25 and below, which decreases as the I:E ratio increases. The problem does not lie in the stepper motor’s rotational speed, as it is able to reach a maximum of 400 revolutions per minute. However, at higher speeds, the motor’s stall torque decreases significantly such that it is unable to simultaneously maintain an adequate torque and RPM to compress the Ambu-bag in the time allotted.

Future iterations of VentCU will likely feature a new motor, one with a much higher stall torque at the velocities necessary for all I:E ratios. Alternatively, it may be possible to achieve these metrics by using the same motor and instead modifying the system to create a mechanical advantage to increase torque.

Pressure Sensor Data Collection

Currently the sensor collects data as expected. However, the pressure sensor readings still must be converted to understood units of pressure and these readings must be verified. This has not yet been implemented. Because of this, the device does not yet have tidal volume measurements. Additionally, alarms relating to pressure measurements cannot be triggered.

PCB Development

To eliminate the need for soldering, we determined that a PCB (printed circuit board) could be used to simplify the electrical architecture and increase the convenience and accessibility of the machine. Ideally, this PCB would contain the ADC (such as the TI ADS1115 chip used in both ADC options), the pressure sensor, and a row of screw terminals for pin access. Having the pressure sensor and the ADC pre-assembled would eliminate the requirement for a soldering iron and would reduce the overall complexity involved in wiring the system. Schematics and related files are forthcoming and will be included in the

GitHub⁠

repository.

Battery Addition

Currently the device uses a wired connection for power. However, after speaking with a number of senior New York hospital clinicians we determined this to be dangerous in the case of power outages or other emergencies. We recommend expanding the electrical system to include a battery capable of powering the ventilator for a minimum of 30 minutes. Disconnecting a ventilator increases the possibility of contagious spread, so when patients must be transported through the hospital it is preferable to leave the ventilator in place during transit. As seen in the diagram below, a computer battery backup can easily be inserted in the bottom level of the ventilator.

⁠

Control System

From a meeting with the anesthesiologist who is leading New York Presbyterian's ventilation strategy, we learned that a number of sensing and control necessities are missing in our current design.

At a minimum, a field-usable ventilator must be able to measure Peak Inspiration Pressure, set the Tidal Volume, set the Inspiration/Expiration Ratio, and set the Respiration Rateーthese are currently part of our design. However, the ventilator must be also able to detect Positive End-Expiratory Pressure (PEEP). COVID-19 patients are particularly susceptible to the danger of auto-PEEP, a condition in which the lungs fail to fully exhale the total ventilator-delivered tidal volume before the next machine breath is delivered. If this situation goes unrecognized, the peak inspiration pressure and total volume pushed into the lungs can raise to dangerous levels, possibly leading the lung to rupture. To detect auto-PEEP, the ventilator must have a more complex flow profile which includes inspiratory and expiratory pauses, allowing the pressure in the lungs to equalize and to measure plateau pressure and PEEP respectively.

⁠

(source:

https://veteriankey.com/advanced-mechanical-ventilation/

⁠

)

More drastically, we learned that sedating patients enough to fully suppress their brain's signals to breathe is extremely difficult, and appropriate medications are already running in short supply. For this reason, our previous assumptions of having an intubated, completely sedated and paralyzed patient may be unreasonable. Continuous Mechanical Ventilation (CMV), the term describing our current strategy of un-triggered ventilation, can cause significant damage to the patients lungs if they are not completely sedated, especially when their lungs are already damaged. Most Bag Valve Masks (BVM's) include Expiratory Valves to allow outside air in during the event of a spontaneous breath, but medical professionals conveyed that these valves are not completely reliable.

The only true solution to the problems presented by CMV is to implement some variant of patient triggered ventilation or assist control ventilation (ACV). This is a far more complex control problem, but a pressure triggered ventilation system, where breaths are triggered by pressure deviations caused by a patients intent to breathe, is not beyond reach with our current sensor suite.

Next:
`Acknowledgements`⁠
|
`Overview`⁠
⁠

Want to print your doc?
This is not the way.

Try clicking the ⋯ next to your doc name or using a keyboard shortcut (

CtrlP

) instead.