Skip to main content

Community spirit goes viral

The pressure of competition can be great motivation and motor of innovation. On the other hand, it does suppress the potential of swarm intelligence. If scientists like Robert Koch and Luis Pasteur had cooperated instead of battling against one another, they could possibly have saved many more lives. Or take an example of exceptional success: Wikipedia has proven the efficiency of swarm intelligence and shown that community review leads to higher reliability of information than any editorial staff could produce.

An MIT team has started the community project "E-Vent" to fight the shortage of medical ventilators. This article is a short review of their aims, solutions and problems, and how the community spirit might help to overcome difficulties.

Respiration: a sophisticated biological function

shutterstock_1561133012_bd4ae473a701c7777ba1fa4c4573a4e4ff33bdfb.jpg

Many teams in different countries have been and are involved in constructing open-source ventilators for medical use. Many of them are not aware of the problems you are facing by interacting or simulating such a sophisticated biological function like respiration. Human respiration is much more than blowing air in and out a balloon.

First of all, there is a purpose for breathing: Your body needs oxygen (O2) from the ambient air and needs to get rid of carbon dioxide (CO2). This constant gas exchange is an essential requirement for metabolism. Without O2, your cells are dying (starting with your brain cells). With too much CO2 (also called carbonic acid) your blood gets acid which in turn causes nerval and muscular dysfunctions and sooner or later kills you.

shutterstock_239128069_%5BKonvertiert%5D_da4423de690e79c3c68046a242fe948787a2a0b9.png

Divers are aware of the damage too much O2 causes for your lungs and the brain. Too little CO2, on the other hand, causes an alkalosis. That’s what happens if you breathe to fast (“hyperventilation”). Breathing into a plastic bag in case of hyperventilation does nothing else but increase your CO2 level to typical values again.

Pressure does influence the gas exchange in your lung. That’s why the Mount Everest tourists need to breathe pure oxygen while queuing for the peak and why aeroplanes have oxygen masks falling down in case of a sudden loss of pressure. But extensive pressure can also be a destructive force for the small gas exchanging chambers of your lung called “alveoli”.  If their sensitive walls disrupt, gas can get between your chest and the lung. This compresses the lungs so that gas can no longer be exchanged (called a “pneumothorax”).

Under normal (“physiological”) conditions, inhalation is started by your chest muscles. They extend the lungs so that there is a negative pressure (compared to ambient pressure) inside the alveoli and bronchi (the tubes connecting the alveoli with the air tube (“trachea”). This causes air to be sucked into the lungs. When the muscles release, the elastic structures of the chest and the lung will press the air out again. The tissue of the air passages is flexible, and the diameters of them are elastic. And of course, the overall volume capacity of the lungs is limited. The mentioned parameters cause a typical pressure curve:

shutterstock_137284214_%282%29_8847df6ae23eab25cb77079b5959eff2a9d981bd.jpg

Whenever a machine tries to support or replace this physiological function, it needs to imitate this curve to a certain degree. But everything is reversed: Positive pressure presses the air into the lungs. As long as the lungs and the chest do have enough elasticity, the air gets pressed out for expiration without using the machine.

shutterstock_1645104367_0c7f1868daad08183fbb871884152594e384d957.jpg

I do not want to get too deep into details, but you may imagine that a sick lung (“pathologic respiration”) had different needs compared to physiological breathing. There is also a big difference in supporting spontaneous breathing (“assisted or augmented ventilation”) compared to the respiration of a patient without spontaneous respiration (“mandatory or controlled ventilation”). In this case, the patient is unconscious and will tolerate externally applied pressure while the first will tend to “breathe against the machine”. Therefore ventilators need to be able to detect spontaneous respiration and adapt to the patient’s breathing frequency.

Professional ventilators have different modes of operation (e.g. assisted, controlled, pressure controlled= PCV or volume controlled=VCV or CPPV and many more). They also provide a constant basic positive pressure to “hold open” the airways (called “PEEP” = positive end-expiratory pressure).

Things get even more sophisticated when comparing non-invasive ventilation (using a mask which is placed over the nose and mouth) with invasive ventilation (using a tube inserted into the trachea). A mask is not always wholly sealed, and a device needs to take leakage into account.

shutterstock_337094036_895c4da0ea1751fa5444a84173a1e5712fffbf87.jpg

Can you imagine how sophisticated a ventilator needs to be? It needs to control the pressure curve, the respiration volume, the respiration frequency (resulting in a “minute volume”) and a trained person needs to be able to continually adapt these values to the individual and current condition of the patient.  

Distinguished argumentation instead of populist slogans

Constructing such sophisticated machines do need a development team capable of knowing the biological and medical facts and anticipating application problems with different setups. Although I do appreciate the intention to be helpful of all those people out there who have started open source ventilator projects, I fear that many of the DIY community are lacking those capabilities. End of March 2020 this review has been published: “A review of open source ventilators for COVID-19 and future pandemics”. The résumé is not very encouraging: “The results of this review found that the tested and peer-reviewed systems lacked complete documentation and the open systems that were documented were either at the very early stages of design (sometimes without even a prototype) and were essentially only basically tested (if at all)”.

When I have studied the online documentation of the “E-Vent” (“MIT Emergency Ventilator”) project, I was very much impressed by their scientific approach. While other teams tend to publish their projects with a focus on media reaction, the MIT team describes their work in a differentiated and academic way. First of all, they have declared clear aims of their project. E.g. although articles like this one have insinuated that E-Vent would apply for an NDA approval, they clearly stated that they are not going to apply for because there is no need for. That said, they do focus on fulfilling all NDA requirements for emergency ventilators. They also emphasise that this is not a project for typical DIY people to undertake because such a device needs to work to be manufactured according to FDA requirements (NHS in the UK). The “determination of minimum requirements was made by a team of physicians with broad clinical backgrounds, including anaesthesia and critical care”.

The US FDA, as well as the UK NHS, do have released official requirements for emergency ventilators:

Enforcement Policy for Ventilators and Accessories and Other Respiratory Devices During the Coronavirus Disease 2019 (COVID-19) Public Health Emergency

Specification for ventilators to be used in UK hospitals during the coronavirus (COVID-19) outbreak

The MIT project team has published a risk analysis for their approach to be discussed by the community. I love this kind of professionality!

The MIT team perfectly summarises the functional requirements with this statement: “We recognise, and would like to highlight for anyone seeking to manufacture a low-cost emergency ventilator, that failing to properly consider these factors can result in serious long-term injury or death.”

Another thing which impressed me deeply when reading the MIT project’s pages was their professional focus on non-functional requirements:

The BOM and manufacturer skills: Every part which is used to build the E-Vent device must be readily available in large volumes. The manufacturing processes must be easy to be setup.

The lifetime of mechanical parts is critical: During three weeks of ventilation, the device works 24/7 and will cycle more than 1 million times. Every material used is checked for this extended lifetime requirements (e.g. no aluminium gears).

shutterstock_1673318152_09ceb6ca4491d622cf47f11934480ab1b9386384.jpg

The MIT team’s approach is to use a manual resuscitator, called “Ambu” bag (also known as “manual resuscitator”). Such bags are used by paramedics in emergencies when re-animating patients. There can be a mask attached to it (“BVM”=bag valve mask), or it can be attached to endotracheal tubes. They are affordable, and millions of them are ad hoc available.

The E-Vent project uses an electronically controlled motor which drives a gear to move two levers compressing the bag. The drive squeezing the bag is part of a closed control loop, including sensors.

MIT-Ventilator-Project-01_1_6caecdcc2c31b8e2961d7a0239b5483bb024677e.jpg

courtesy of the researchers, MIT

The team has rejected other approaches like using CPAP devices used for sleep apnea therapy. Medical experts agree worldwide that ARDS (“acute respiratory distress syndrome”, the syndrome caused in severe cases by COVID-19) cannot be cured without invasive ventilation (“intubation”). So mask ventilation could only be a temporary solution for a hypoxic respiratory failure.

shutterstock_4278678131_cd35ed79149b62344e1fff1e8ffab0ce36c6615d.jpg

Many other teams who have decided to go this way have not carefully thought about this problem: Their solutions risk aerosolising COVID-19 in the expired air. This is a high risk for the medical staff, especially in times of shortages of protective material, and it must be avoided. The MIT team has recognised this risk.

Using swarm intelligence to solve problems

The E-Vent project has faced many challenges. Some of the problems are still not solved. But they have not only decided to be an open-source project to share perfect results. They also hope for the worldwide community to comment on their thoughts and to help with problems.

Let me address just one of these challenges which is often not even considered by other teams:

When I was a kid, I thought I would have invented the coolest idea ever: I wanted to put a two-meter hose onto my snorkel so that I could dive for hours below the water surface without risking to get water into the snorkel. Without a wise older adult telling me I would risk my life, I probably would have ended as fish food. The problem is called “dead space”: The expired air is blending with the fresh air inside the hose. Re-inhaling this mixture, again and again, reduces the O2 concentration, and what is even worse it increases the  CO2 concentration way above the toxic level.

With a ventilator, you need to take care of a similar problem. The dead space must be as small as possible. A manual resuscitator has an “exhalation port” directly attached to the “patient connection port”.  Under usual conditions, there is only a very short hose between the endotracheal tube and the patient connection port. A two-way valve (“patient valve” controlled by the pressure difference between bag and lungs) allows airflow either from the lungs to the exhalation port or from the ball to the lungs. There can be a PEEP valve attached at the exhalation port:

BVM_5111ad5526b8c57722dab9f5c01a9a8958fad3fe.png

That means, during expiration “used” air (with high CO2 concentration) is not getting in any larger space, but the patient valve and a small piece of hose and is directly lead into the ambient air.

But with the E-Vent device, things are different: There is a larger hose between the patient valve and the endotracheal tube. This hose is drastically increasing the dead volume, which could be a high risk for Co2 intoxication. The team says: “in a 1 m long tube of nominal 2 cm diameter, there is an unacceptable 314 mL dead space that the patient will breathe in and out and not be oxygenated.” And “A way to move the patient valve of the manual resuscitator closer to the patient is critical in solving this issue. Standard ventilator circuits have two limbs, one for inspiration and one for expiration so that gases can be recaptured by the ventilator. Single limb ventilator circuits with a patient valve located distally already exist on the market, but are not necessarily optimized for use with a manual resuscitator. Solving this problem requires creativity – to the best of our knowledge, no manual resuscitator manufacturer makes an approved solution, and no manufacturer makes all the parts that will assemble together correctly.”

This and other problems need creativity. And the combined creativity of many people can be powerful. The MIT group wants people to join their online discussions. And if you have a look at the page dealing with that problem, you will see that there is already a potential solution. They say: “Constructive comments are welcome from any and all individuals who would like to contribute; this is essential to ensuring a worldwide team effort.”

Get involved!

I’m more than happy to get the chance to write for so many creative readers. And I’m sure that some of you 800,000+ registered DesignSpark members of the DesignSpark family will be able to add great comments and ideas to the E-Vent project.  Please go ahead and visit their project at

https://e-vent.mit.edu/

Your input to this project could help save lives. And of course, they have provided a link to support them financially. DesignSpark will follow their activities and inform you as soon as there is new information about the E-Vent project.

Let me close this article with a citation from the E-Vent project’s web pages:

Finally, our thoughts are with all the healthcare workers on the front lines and the patient’s battling this illness, those keeping grocery stores, pharmacies, and critical infrastructure functioning worldwide, and everyone who has been affected by this pandemic.

 

Volker de Haas started electronics and computing with a KIM1 and machine language in the 70s. Then FORTRAN, PASCAL, BASIC, C, MUMPS. Developed complex digital circuits and analogue electronics for neuroscience labs (and his MD grade). Later: database engineering, C++, C#, industrial hard- and software developer (transport, automotive, automation). Designed and constructed the open-source PLC / IPC "Revolution Pi". Now offering advanced development and exceptional exhibits.
DesignSpark Electrical Logolinkedin