Over the last several weeks GasLab has received dozens of calls from engineers and universities trying to develop low-cost oxygen ventilators to assist patients infected by the Coronavirus. After hundreds of emails and phone calls - as well as doing our own research into this topic - here are some of the things we’ve learned. We hope this is useful to others starting their own ventilator project.
Mechanical Resuscitator Bags as Ventilators
Social media sites like YouTube show several examples of manual resuscitator bags being mechanically “squeezed” to act as a ventilator. The bags force ambient air (21% oxygen) into patient lungs to assist breathing. While manual resuscitator bags can be used for a short time, they are not able to control air volume, pressure, rate and flow.
Coronavirus patients may be on a ventilator for days or weeks. If a mechanical resuscitator bag is used for that long it can cause barotrauma (damage) to the alveoli in hours due to the increased air pressure inside the lungs. Therefore, adding supplemental oxygen or oxygen sensors to a mechanical resuscitator bag is not the most important issue that needs to be addressed. Bio-feedback to control breathing rate, air volume, pressure, flow and humidification are as important as supplemental oxygen in a ventilator design.
Those looking to design a mechanical resuscitator bag should first watch this video: “A Guide to Designing Low-Cost Ventilators for COVID-19” by Real Engineering. It gives a good overview of best practices and pitfalls when considering a mechanical resuscitator bag design.
Another source of good information about the various groups looking to make ventilators is the Vice.com article "People Are Trying to Make DIY Ventilators to Meet Coronavirus Demand."
Recent Ventilator Projects
One of the best potential solutions to the ventilator shortage could be the AfVa Advanced Ventilator already being made and being sold in India. While not used in the US, the advantages are that it uses a free smartphone app to control the machinery and only costs around $2,000 per unit.
A US company MainGear.com has built the LIV ventilator using 3D printing and off-the-shelf parts. They say the price is around $7,000 per unit.
For anyone interested in building a ventilator you should also read this article "If You've Heeded the Rallying Cry for Ventilators Here are the Government Regulations You Should Know." It gives a list of best practices for electrical engineers as well as discussing the relaxation of government standards in ventilator design.
Oxygen ventilators like those used in hospitals combine the mechanical ventilator action and supplemental oxygen.
Oxygen ventilators use more than a mechanical squeeze bag. They control the air volume, pressure, rate and flow in part using a feedback loop from the patient’s own breathing.
Because patients with lung damage may not receive enough oxygen into the blood from ambient air, oxygen ventilators use pressurized tanks of pure medical-grade oxygen mixed with filtered and humidified ambient air to supply an oxygen-rich mixture of air to the patient. Depending on the severity of the trauma, up to 100% oxygen may be required.
Before designing an oxygen ventilator it would be helpful to watch this video: “Mechanical Ventilation Explained Clearly” at MedCram.com. This series of videos was created for medical personnel to understand the workings of a ventilator, but it has been made freely available to anyone interested in the medical requirements of oxygen ventilation.
Medical Oxygen Sensors
To control and monitor the oxygen level of a Coronavirus patient on a ventilator, sensors are used to measure oxygen both in the gas being given to the patent as well as the exhaled oxygen level - also referred to as end-tidal oxygen (ETO2).
For oxygen control, ventilator manufacturer’s typically use FDA-approved electrogalvanic sensors, sometimes referred to as "disposable oxygen sensors.". These relatively low-cost and dependable devices measure 0-100% oxygen levels.
The downside of electrogalvanic sensors is that they are slow and short-lived, often lasting only 6 months. They can be poisoned and are cross-sensitive to a large range of gases. In addition they are thermally and pressure dependent so they need several different compensations. However, they are simple to operate and have been designed over a long time to fit the needs of oxygen ventilator manufacturers.
Some manufacturers also offer 0-100% medical-grade electrochemical oxygen sensors. While electrochemical sensors are not as fast as electrogalvanic sensors, they are low-cost, have relatively longer lives (2-3 years) and higher accuracy. Like electrogalvanic sensors they are biased by humidity and pressure and are cross-sensitive to other gases.
Both electrogalvanic and electrochemical oxygen sensors can be found by searching for "medical oxygen sensors" on Google. If you want an overview of several different sensor manufacturers products visit ORSupply.com or Sensoronics.com.
The challenge for anyone interested in creating a new ventilator using a current medical-grade 0-100% oxygen sensor is:
- Most of these sensors are made to purpose for a specific ventilator. Documentation and tech support for the sensor itself (power requirements, output signals) is difficult to find online.
- Many of these sensors are in short supply. It makes no sense to create a new ventilator using an oxygen sensor already in demand during the Coronaviris crisis.
This is why engineers are exploring non-medical oxygen sensor alternatives.
Non-Medical Oxygen Sensors
The oxygen sensors we stock fall into 3 classes: electrochemical, zirconia and optical. Each have their strengths and weaknesses as well as suitability for purpose. While not FDA-approved for medical devices, they are being used to test proof-of-concept or to move preliminary designs forward while a production oxygen sensor is being sourced online.
Electrochemical Oxygen Sensors
The electrochemical oxygen sensors offered by Gaslab are typically used to measure oxygen levels in ambient air. They are used in products like hand-held gas detectors. Although their low cost and high availability makes them attractive alternatives to medical oxygen sensors, they only measure up to 21% oxygen. Therefore, they are not suitable for controlling ventilator oxygen levels up to 100%.
Zirconia Oxygen Sensors
While GasLab doesn't offer automotive zirconia-based oxygen sensors, they are the most common type of oxygen sensor sold worldwide. We get asked about them regularly. Automotive oxygen sensors measure the oxygen burned during combustion. This data is used by the auto’s computer to fine-tune the air-gas ratio for maximum fuel economy.
The problem with automotive zirconia-based oxygen sensors is that they are tuned to work at one specific oxygen level, not a broad range like 0-25% or 0-100%. Therefore, automotive oxygen sensors are not suitable for oxygen ventilators.
However, 0-100% zirconia oxygen sensors do exist. For example, several people have purchased our SST Zirconia Oxygen Sensors or our Fujikura Oxygen Sensors to develop proof-of-concept products. They are accurate, dependable and have a long lifespan. Unfortunately their relatively high price makes them impractical for low-cost oxygen ventilators.
Optical Oxygen Sensors
Our most popular optical oxygen sensor is the LuminOX LOX-02. While only a 0-25% oxygen sensor, it’s low cost, high speed, accuracy, USB devkits and our free GasLab software has made it popular for engineers who need to quickly measure ambient oxygen levels. However, because it cannot measure up to 100% oxygen it is limited to devices useful for measuring ETO2.
Another type of optical oxygen sensor we offer is the 0-100% Optochemical Oxygen Sensor by TecSense. These sensors are rugged and stable, have no calibration requirements, cannot be poisoned, and are very fast and accurate. However their high price makes them too expensive for ventilator applications.
We’ve spoken to many engineers who say they don’t know anything about ventilators, but their company or university has tasked them with creating a low-cost ventilator to help hospitals during the Coronavirus crisis. We encourage them to work with a pulmonary specialist early in their design and to come up to speed on the medical requirements of a ventilator before they focus on the mechanical design.
Oxygen level sensing is minor when compared to the pressure and flow monitoring and control that is required. Most have not investigated the ETCO2 and capnography aspects, or condensate management required for oxygen ventilation.
The good news is that with some of the best minds in the world working on this important project, we’re confident a new generation of low-cost, easy to build ventilators will soon be on the market.