| September 28, 2023
Robert Read lead the development of a ventilator test device during the Covid19 pandemic, it became the Ventmon.
By the Open make team and Robert Read. Copyright to the authors, distributed under a CC-BY 4.0 licence.
Banner image: VentMon logo, distributed under a CC-BY-SA 4.0
Interviewee: Robert Read
Interviewers: Robert Mies (TU Berlin) & Moritz Maxeiner (FU Berlin)
Transcription and editing: Diana Paola Americano Guerrero, Robert Mies, Fabio Reeh, Moritz Maxeiner & Julien Colomb
Screenshot of the interview.
Photos of the VentMon
- Main URL: https://github.com/PubInv/ventmon-ventilator-inline-test-monitor
- Project start: 2020
- Core development team size: 4
I built a ventilator test device which became the Ventmon. It has wifi connections with “Internet of Things” enabled.
It allowed a quality assurance for the teams that were attempting to build mechanical ventilators.
The VentMon is halfway between a demonstration and commercial product, it is the most mature product. You can’t buy it right now and I don’t have an inventory of them.
The PolyVent is somewhere between a prototype and a demonstration because we attempt to build an educational platform with it.
It was built by the team and used to test ventilators worldwide.
Would you call yourself the principal investigator in the VentMon project ?
The VentMon project started in March of 2020 when COVID-19 was first identified. There was an extremely urgent belief that there might not be enough mechanical ventilators for all the people.
Instead [of building a ventilator like other teams], I wanted to be in service to those teams. I knew they all needed a test device. Instead of making a ventilator, I built a ventilator test device.
How did it all start with the VentMon project?
Could you say what have been the core benefits of this project? How does the hardware fit into the overall project?
It was funded by my nonprofit, which is called Public Invention, itself funded by donations.
We got grants to manufacture and give Ventmon devices away. One was from the Mozilla Open-Source Foundation, and one was from Protocol Labs.
How is it funded?
On this project I got lucky, and I had really good volunteers and that wasn’t a problem.
We tried to identify libraries or sub modules which can be taken out and reused.
I’m taking techniques developed by the open-source software world and applying them to this new realm.
Not all of our work is currently documented at good level because the documentation often kind of lags behind where we are. The goal is to document everything with CAD files, Bill of Materials and tips like use a hot melt glue gun to attach this to the side of the enclosure, those kind of assembly instructions which are necessary for someone to reproduce the device.
Could you describe the overall process how that happened?
What major issues have you come across during the project and how did you resolve them?
Did you have problems on a technical level?
One question about the time series part, do you know why it isn’t common to put datetime timestamps?
We made decisions as a team with the four major volunteers Lauria, Geoff, Ben, and myself: a small team.
How were decisions within the project made?
There is a software system that’s very important. The idea was to build VentOS and then we use hardware drivers to change based on which kind of mechanical equipment you have. VentOS GUI which is closely related to VentDisplay, is a software control mechanism that’s like a clinical display a doctor would use.
Which hardware projects have you developed as part of this project? Just the ventilator or are there more components?
One thing we did that I’m very proud of was define a data interface called the Public Invention Respiration Data Standard. But I have not yet gotten anyone who isn’t directly associated with my teams to use that data standard.
The technique may very well have been invented 100 times, but no one has taken the time to produce an international standard for it. One team had their own data standard that was different from mine. I wrote an adapter for my data standard into their data standard.
It doesn’t matter how reliable the devices are, they must have alarms: Things go wrong even if you built the very best device in the world. Right now, the PolyVent has no alarm mechanism, it cannot make a noise or a light. I’ve been trying to recruit volunteers to make a general-purpose alarm module.
Most people would say the best I did was the spreadsheet which was a service to the community: it made something that was invisible visible and discoverable.
Are there other parts that need to be developed?
Where would you rate your products and the components on a scale of prototype, demonstrator, market-ready product?
Are you aware of anyone who tried to reproduce your hardware from the online schematics and information?
What was your goal of the publication? Do you want people to rebuild it, or did you have other goals?
Has there been any further development or contributions you found?
Everything we do has an open-source license on it from the start.
They’re a technical report, from an engineering point of view on how to build something, they’re not the kind of thing that is considered to be interesting from an intellectual point of view.
That’s for the peers to say not for me to say whether it’s worthy of publication or not. If it doesn’t, it will be a non-peer reviewed publication at GitHub.
It’s a lack of time that keeps some things from being published.
What are the envisioned outputs of the hardware development for all the parts on the components?
I like to write academic papers. The VentMon has been published in Hardware X. I believe that is a useful way to disseminate knowledge.
About the contents of what you’ve shared on GitHub and within the papers, are there Bill of Materials, CAD files, assembly instructions, documentation? What does it all entail?
Do you intend to publish those components in Hardware X if you have enough time?
How did you come about GitHub and Hardware X and why did you choose those platforms?
Have you encountered any barriers in using either of those platforms that you had to struggle with?
Do you have other reasons of not publishing results that you’ve not published yet?
The goal of Hardware X is that anything could be reproduced by an average engineer. Would you say that your projects have reached that stage and aren’t published on Hardware X yet? Or is there still work to reach that level of reproducibility?
What’ve been the successes you reached in this project? At what parts you have not succeeded?
You said at the Open Hardware summit that the success rate of all the projects is very low because most of them haven’t made it to the market. Therefore, your list may not show a great success in terms of impact in reality.
I believe in a coming Star Trek universe, a bright universe where humanity works together, national differences have dissolved, and we do wonderful things.
I didn’t do this until I became financially secure. Now, I can finally do what I’ve always thought should be done. I started this nonprofit.
All of these people did it out of the goodness of their heart. All of them don’t mind getting academic papers. Geoff doesn’t need an academic paper, but Lauria’s career was advanced by an academic paper. Ben Coombs finds the work interesting, and it expands his universe, kind of socially.
For young people, I think they had a very strong resume type situation. I will write them a strong recommendation.
How did you end up working on that project? In fact, what motivated you as a person how did you decide for it or what made you start?
Could you tell us about the other people you’ve worked with? How many members worked on the project? What’ve been their occupations? It’s not just doctoral researchers, postdocs or research engineers, could you break it down?
What I can do with my technology, is try to build the best technology that helps people instead of hurting them. That is what motivated us.
No one, including me, gets paid by Public Invention.
How did you find those people?
Many of the teams using the VentMon were geographically distributed.
We held some virtual symposia and workshops.
We use GitHub and Git issues to represent issues and weekly meetings. I’m a big proponent of agile development. When I was in industry, I was a manager who used user stories, velocity tracking and the whole scrum framework. I know how to do that. I find that people don’t want to do that.
How did you coordinate the work with these people?
How have the members benefited from the work in the project?