Hacker News new | past | comments | ask | show | jobs | submit login
Glia – High-quality low-cost open source medical hardware (glia.org)
595 points by robtherobber 33 days ago | hide | past | web | favorite | 50 comments



They are certainly working designs and products, but I am not sure there are actually medical devices[0].

The research (self) validation [1] talk about similar performances than an existing product.

However, it does not talk about bio compatibility of the material used, potential contamination during manufacturing process, what kind of disinfectants can be used on the product, what is the operating range in terms of pressure, temperature, humidity, etc

A proper medical device manufacturer is also supposed to have a complaint handling unit, to raise and analyze every adverse effect on the field, and report to regulatory bodies the worst hazardous case in a limited time frame, be able to do recall and so on.

So I believe that this coming from the Gaza strip is a real boon because you don't have any other choice, but it does not mean that you could translate that model everywhere in the world.

Edit :

it seems this is registered as a Class 1 medical device in Canada, so it is possible that all my previous point have been addressed (still I am not quite sure how you can validate only a design without the manufacturing/shipping part) I am working on a Class 3 device so it is possible I am more paranoid that I should be...

>> In Canada, it is produced by Glia as a Class I device. Glia holds a Medical Device Establishment License from Health Canada.

I am not sure how you could pass an audit if your documentation contains "Lorem Ipsum" [2]

[0] The homepage mentions medical Hardware, but this page https://docs.glia.org/ still talks about medical device

[1] https://glia.org/stethoscope/

[2] https://docs.glia.org/docs/stethoscope/manufacturing/


I think the attitude you approach this company with highlights one of the major causes of the unjustifiably high medical costs in the US. Many of your critiques are based on status quo expectations that are, for the most part, non-issues.

Yes, these devices are of a lower quality than devices with prices 10-100x higher, but they get the job done, which is what actually matters. Some new practices will have to be designed to properly decontam and maintenance them, and experimentation to figure out their tolerances, but there are many more medical practices in the world that need these devices than have them; the main issue is cost.


"However, it does not talk about bio compatibility of the material used, potential contamination during manufacturing process, what kind of disinfectants can be used on the product, what is the operating range in terms of pressure, temperature, humidity, etc"

Working in hospital infection control, this was one of my first concerns as well.


> I am not sure how you could pass an audit if your documentation contains "Lorem Ipsum" [2]

In a lot of industries there’s a huge conflict of interest as the auditee is also the company paying the auditors.

If the auditee loses a certification / accreditation, that’s also a loss of business for the auditor.


> If the auditee loses a certification / accreditation, that’s also a loss of business for the auditor.

A bunch of simple medical devices (Class I in the EU) are self-certified by manufacturers, and require no seal of approval before commercialization.


Because they can cause no harm, otherwise they would be in Class II.


Does US class 1 require - "complaint handling unit, to raise and analyze every adverse effect on the field, and report to regulatory bodies the worst hazardous case in a limited time frame, be able to do recall and so on." - Don't know so I'm asking. I would assume any product with class 2 has these requirements. My assumption under class 1 is that you need to be able to determine you built the product to a predefined set of specifications.


The device is registered in Canada and I found that information [0] :

>For all devices except Class I, implement an ISO 13485:2016 under the Medical Device Single Audit Program (MDSAP) compliant quality management system, which includes the specific requirements of the CMDR. ISO 13485 certification, used to demonstrate compliance with European regulations, does not meet Canadian requirements. Updates to the existing procedures, or new procedures, must be implemented.

So it seems you don't need to implement post market surveillance for a class 1 device or having a complaint handling unit.

[0] https://www.emergobyul.com/resources/canada-process-chart


Yes. Complaints are part of the Quality System Regulation that all device manufacturers selling in the US must abide by.


The devices they currently market are low risk (across US, Canada, EU, etc) so there is minimum regulatory burden. The "working on it" note regarding a Pulse Oximeter mentions a clinical trial, which I infer to mean that they're considering regulatory requirements for higher-risk or more complex devices.


This is certainly cool and I love my Craftbot 3d printer. However, one has to mention the problems with using 3d printed devices like this. This is not to tear apart the project, but I would hope valid criticism. Medical gear like this is commodity stuff that is actually very inexpensive. I do think 3D printing will democratize the really expensive medical equipment, but we are not there yet.

#1 cleanability - It is very hard to properly clean a 3d printed item and each print may be a little different than the others.

#2 Cost - For the same cost, I can buy a stethoscope on Aliexpress that is made from aluminum and pvc. For about $5 I can purchase an actual Littmann replica made from the same durable parts as the previously mentioned steth from Aliexpress. The same goes for all of the designed products they have.

#3 Durability - PETG and ABS are relatively durable, but they do not compare to aluminum in terms of strength. In emergent situations, I would not trust plastic for my gear. In the clinical setting, I can see the use case, but still like the term "buy once, cry once" when it comes to tools that will see lots of use.


I actually attended a talk by Dr. Loubani this past weekend at MIT and one thing he emphasized was that everything they do has to be reliably replicated in Gaza. So yeah, one could conceivably purchase the Littmann replica from AliExpress but getting it into Gaza? Another story...


I figured an interesting model for open-source hardware would be to create nice designs.. CAD models, schematics, gerber files for making PCBs, firmware etc... put it all on Github, and wait for Chinese factories to start producing them.

Here's the kind of device I had in mind.. but maybe there are more immeadiatly useful types of machines to work on, I don't know:

https://www.cmecorp.com/welch-allyn-28600-tm286-auto-tymp.ht...

"Designed to make detection and documentation of middle ear pathologies fast and accurate."

Happy to hack on stuff if there's some people to work with.


Try buying something from Aliexpress when you have a close border with Egypt and Israel, and a blockade within Israeli sea and air.


Good Luck team Glia! Keep fighting the good fight. Most people look at insurance, hospitals, pharmaceuticals as too profit hungry. Medical devices are equally to blame. All the big OEMs have doctors (sales person) selling to doctors in the surgery room so that their products, which haven't been materially updated in 60 years can be sold at a 1000% margin. Its another industry where the price should, in theory, be getting lower, but here we are. Good Luck Guys!


Seeing projects that are neither fun- nor money-driven (though both are valid motivations and doesn't exclude this) but some guys really trying to improve a tiny bit on this world makes me sentimental lately.


I was involved in the early stages of the project, more specificity I designed the 3d printed stethoscope prototype:

https://journals.plos.org/plosone/article?id=10.1371/journal...

Happy to answer questions


What 3D printer and software did you use?


I got a prototype of the Lulzbot mini, which I used to print it (and I still have that printer!).

I designed it in a ruby thing that translates things into openscad (crystalscad was the name, but I deprecated it for multiple reasons. There's jenncad in my github repository as its successor, but I didn't have the chance to get it anywhere to stable)

Other software I used were printrun and slic3r.


For everyone interested in this topic, another non-profit in the same area is Cadus, a Berlin-based makerspace producing innovative low-cost medical infrastructure for crisis response.

https://www.cadus.org/en/


I'm wondering: if we can get a can of Coca Cola to just about any part of the world, then why can't we (with the proper logistics) get basic medical equipment there? And wouldn't this equipment be cheaper and/or of higher quality than 3d printed?


We can. An aliexpress stethoscope is 2.34 with free shipping.


Good luck buying from AliExpress in Palestine


You don't have to buy directly from AliExpress in Palestine. There are thousands of people that resell AliExpress in your common street market. A huge percentage of everything you see street vendors selling is ultimately sourced from China for 2.34. The issue is that every middleman takes a bit of markup until it reaches the street. So now you're paying 10, cash, and you have it immediately.


Not just Palestine, but Gaza. The cost of getting anything, medical equipment especially, into Gaza is insanely high and it's usually an unreliable process. Thats likely why theres such a huge emphasis on being able to create the entire unit in the Gaza Strip.


Payment issues. Sending and receiving money to/from countries deemed high risk is hard. Can't do a normal bank transfer to Venezuela, for example. That applies to Coke, too - "we" can't actually send one to any part of the world.


why can't we? Politics.


The tourniquet is a "knock-off" of the North American Rescue CAT[1]. I'd be pretty wary of using it in real life... The windlass ends up under a lot of shear force, and I would be surprised if a printed thermoplastic was able to deal with that without failing with some regularity. I've seen cheap injection molded versions snap, and I'd expect the same here.

In reading their "field validation" blog po... err... "reports", it looks like they failed pretty badly. Seems irresponsible to be publishing a design they know is badly flawed.

[1] https://www.narescue.com/combat-application-tourniquet-c-a-t...


They fixed those flaws, the second deployment of 78 units had no failures (content warning: this post is somewhat graphic)

https://medium.com/@trklou/3d-printed-tourniquet-day-2-of-ga...


I don't put a lot of stock in the second test. Assuming the same people were involved, it's likely they were (consciously or not) being a lot more gentle with the devices.


I was one of the medics there. The failures have stopped after the modification, even with new medics. One of the major issues is that it's such a chaotic scene that it is essentially impossible to fine-tune the stress applied to the device, so your specific concern doesn't apply here. Our testing on the bench also showed that the refined windlass (rod thing) did not cause the same failure even when about 10x the force was applied.

We have received ethics approval for a proper (unblinded) head-to-head multi-site RCT between our 3D printed version, our desktop-injected version and the premium brand, but it will take time for that data to shake out.

tarek : )


AFAIK they had better reliability than regular ones.


I'm sorry, how is that post pretty graphic?


Updated to say "somewhat graphic" because you're right it's not too bad.


Honestly, I found the image of a boy in pain, shot with something that teared up his calf pretty horrific.


There's also the photo of the dead paramedic, which while it isn't explicitly graphic, it would certainly be reasonable for someone to find it upsetting.


Off topic, but I knew that I recognized the guy in blue scrubs on their home page! Tarek Loubani is a Canadian doctor, who made national news when he was shot by an israeli sniper.

https://en.wikipedia.org/wiki/Tarek_Loubani


What's interesting is none of these people are in the US. I feel that the regulatory burden, pay scales, and the political climate are shifting a lot of the scrappy innovation to outside of the US.

ed: awesome ideas, I hope to see tons more designs delivered.


This is awesome! I think these initiatives are urgently needed, especially in the case of pulse oximeters and electrocardiograms.

For both, the components (of medical quality) are extremely cheap, but there aren't any clinically validated open source devices in the market. I think they're relative low-hanging fruits in biomedical engineering, and they have a lot of impact potential.

I really think open source medical devices are the future, I would love to see more biomedical engineers and doctors excited about this.

Does anyone know if there is a roadmap for their pulse oximeter/ECG? I would love to collaborate remotely.


I'm just wondering regarding the stethoscope, I was looking at - https://en.wikipedia.org/wiki/Stethoscope#Acoustic which mentions there are seem to be two types of acoustic pickup, bell or diaphragm. The Glia one appears to be the bell type I think?

It says that the "bell transmits low frequency sounds, while the diaphragm transmits higher frequency sounds". Is there an advantage of one type over the other, or are they used in different scenarios?


Hey, I made the prototype for it. I reverse engineered a Littmann cardiology 3 and used a similar diaphragm setup. For the early prototypes I used the one that came with the Littmann stethoscope; later on we got the same result with different approaches, such as cutting out one out of transparent plastic sheet or using shrinkwrap around it.


Cheers, that's very interesting!


They're used to listen to different things. The diaphragm is in my experience much more important. TBH I never even use the bell. Some pediatricians like it for infants because it is physically smaller (but they put a smaller diaphragm over it). Lots of stethoscopes don't even have one.

Funny article about the bell: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1790765/


The usual answer to such question is "it depends".

That being said, heart related sounds are generally low-frequency while breathing/lung related sounds are broadband so they do contain a significant higher frequency part.


I'm curious about their dialysis system. I'm presently on dialysis myself. Some of the technology is straightforward. Its mostly making sure everything is sterile or you get a massive infection. The simplest technology is literally a bad of sugar solution that is fed by gravity and a second bag to drain out the solution. No power needed except to heat the solution to body temperature but you can manage without heating in a pinch. Again it all has to be super sterile so lots of one time use of plastics.


What are you wondering about specifically?


Cool! Now it is time for you (alleged hackers) to produce cool software for these devices, open sourve preferred. The heart is the most critical muscle of the body. It is possible to detect a whole lot of disease using ecg data. Also there is alleged tibetian pulse diagnosis and where are some papers about pulse diagnosis.


This seems really cool in theory - my main question, though, is why does a Tourniquet need to be 3d-printed? Cant one make one from a cloth and a piece of wood?


A startup that is focused on real issues besides games and social networking (aka advertiser playgrounds) — give these guys some props. Coincidence it didn’t come from the Bay Area?


This is awesome. We seriously need affordable healthcare so it's great to see people innovating in this field to bring costs down.




Applications are open for YC Summer 2020

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: