Hemdeep (00:13)
Welcome to Big Ideas in Microscale, the podcast where we explore groundbreaking research happening at the microscale, where micro innovations makes a big impact.

Hemdeep (00:23)
We're excited to showcase the incredible work being done by our users from around the world who are pushing the boundaries of microfluidics, lab on a chip, organ on a chip, and beyond. Through these conversations, we hope to learn from their experiences, uncover their insight, and bring their big ideas to wider audience. So whether in a lab, on the go, or just curious about the future of microtechnology, join us as we dive into big ideas at Microscale.

Hemdeep (01:08)
Welcome to Big Ideas at Microscale. I'm your host, Hemdeep, co-founder of Creative Cadworks, Cadworks 3D, and Resinworks 3D.

Robin (01:19)
And I'm Robin, co-host and also the technical writer within the marketing team.

Hemdeep (01:24)
Today's episode is part three of our chat with Irwin and Philippe from the University of Cambridge. We've been diving deep into their research on 3D printing for cochlear models and their impact on implantation procedures. If you missed those episodes, make sure you catch up on their incredible journey thus far.

Robin (01:42)
Today we're talking about their future research direction and about something a little different. The company they founded called COSA LTD, which is revolutionizing surgical training and anatomical modeling. But that's just the beginning. So stick with us as today's episode is packed with exciting developments and new directions you won't want to miss. Let's jump into a big ideas at Microscale.

Hemdeep (02:06)
So, and now that you've got all of this work behind you and you have brought solutions to the table, what are the next steps in terms of, you know, the current methodology that you've developed in order to evaluate a cochlear implants? What's the next step for you guys in terms of research? Give us some idea of what the next steps are in front of you.

Iwan (02:29)
Yeah. So I guess from, from the research perspective, we're really looking at having individual cochlea that we manipulate the shape of. And now we're looking at really expanding that into one that like all the different shapes and sizes that we can get our cochlea from that kind of database that we've segmented.

As well as looking at other factors for like the surgical considerations. looking at the surgical drilling site and how you can actually effectively insert the, the cochlear implants. So as well, then they're looking at trying to improve the cochlear on a chip and really take a lot of those learnings and combine them with some of the computational models that we've have. So you can understand from each different type of model.

you have its limitations, be it computational, be it physical or the cadaveric kind of real tissue. It is usually a trade off of the amount of detail you can interrogate and actually measure the model versus the accuracy of the model. So validation is kind of key.

And started a small company, COSA, LTD, that really spun out from a lot of the expertise that we generated from this. And we collaborate with some of the Cochrane plant companies, different clinicians, et cetera, through our work and through those connections, we kind of develop these different projects together. And we're actually getting some traction on that side to look at all sorts of different things in both the realm of the year or mostly around of years, be it more educational.

visual models or R &D projects to look in detail how pressures forces and things vary for medical device development and drug delivery, for example. But then we're also looking at the making really realistic phantoms more generally as well for head and neck to start with and a little bit of everything. We've even printed a mouse for a local startup who wanted to.

We're developing different devices to look at mouse tumors. So, a little bit of everything.

Hemdeep (04:26)
The current work that you've done and the procedures and the papers that you've put out, do you see this being applied to other applications? There's cochlear implants. Are there any other implants or are there any other surgical procedures that you can see that this type of procedure could make it more successful, provide a unique research point for others as well?

Filip (04:49)
What is at the moment rising quite a bit is the drug delivery and treating the cochlea or the hearing with drugs instead of cochlear implants. So cochlear implant is sort of like a universal solution as far as you have auditory nerves in cochlea functioning and you can neurons there. You can most likely stimulate them using your cochlear implants. But the problem with cochlear implant is that first of all, it can damage the residue hearing.

And second of all, it's ⁓ permanent. So it will be there for forever. Some diseases and some hearing loss can be treated with injection of, for instance, a protein. So there is a delivery of a protein, is specifically connects, for instance, neural pathways. And by just introducing this protein into a cochlea, into certain regions of the cochlea, the patient can hear again without...

the need of cochlear implant and having a device inside. So now there are companies, pharmaceutical companies, looking at these drug delivery systems. And that also uses an insertion because you need to either insert the drug into the middle ear and then through the round window membrane, the drug gets inside of the cochlea, or you can insert the catheter into the cochlea itself through the round window as you do with the cochlear implants.

and inject the drug this way. Each technique has its advantages and disadvantages, but this is how you can leverage our research, let's say, the insertion forces to optimize the insertion of the catheter into the cochlea.

Iwan (06:32)
Yeah. The professor that we work with, Professor Manahar Bansi, has the claim to fame of being one of the first surgeons to perform a gene therapy in a year, which is really exciting. And actually there've been quite a few news features about it to show the first patients who actually have like really great results coming out from that. these gene therapies have the potential to be curative. So especially for very specific.

genetic hearing loss, then they're really exciting in the way that things are going.

Hemdeep (07:05)
And the research that you guys have done in terms of the insertion forces would be the one that would be the driving factor. Am I correct?

Iwan (07:13)
Yeah, so there you really need to be very careful of maintaining the cell population, right? So you really want to reduce any trauma that you have during the delivery of any like a drug or gene therapy or even cell therapies. There's quite a few companies in that space as well. Making sure that you don't damage the residual tissue is really key to ensure that therapy is successful. And there's a lot of work both on the, not just the kind of

surgical forces itself, its understanding of the anatomy and what are the key features that you kind of extract and the surrounding anatomy as well. So even making kind of surgical training tools to allow the training of these kind of new novel kind of surgical techniques is very important.

Hemdeep (07:56)
And then the second part is this new company that you've started. What is the mission statement of this company? What are you guys hoping to achieve? What is the current research or the direction that you are hoping to move ahead in?

Filip (08:13)
We started this company based on our research and based on the interest of medical device companies in our studies. So basically how this all started is that we mainly create anatomical models and surgical training tools for ENTs and specifically for, in general, for head and neck. So we started with ENTs and with the inner ear anatomy because that's where we have most experience. We are also working on

other projects, including the head itself or neck systems or nose and models like this. couple of exciting projects we are working on right now and we think would be really nice. For instance, drilable temporal bones. So we identified that there might be a need for these drilable temporal bones as ENT surgeons typically need to undergo a couple of these courses and try to

drill the temporal bone, create mastoidectomy and see how they can access the inner ear. The problem with this is that these courses are run only a couple of times a year or over UK and like globally as well. And they use cadavers and the problem with cadavers is they are scarce and you need a certified lab to run cadaveric courses. And so what we're trying to achieve with these models is to create a cheaper alternative.

the drilling so then let's say junior surgeons can practice more on these bones which we optimize to behave as the real bone during the drilling so the feel should be the same. The bones are again coming from micro CT scans so the interesting thing about the temporal bones is that it's fairly porous, there are lots of air cells inside so when you're drilling through you know the feel is different than if you drill just through a solid plastic.

So there was a challenge to replicate these. We are still optimizing these models and what we're trying to achieve with this product is to really create a validated model, which is certified by the ENTs to behave and to be a good replica of the cadaveric models. And again, the advantage of 3D printing is that we can introduce malformations. We can introduce unique cases into each model.

So then every participant of the chorus can practice these unique cases, which can happen one into 10,000 patients, you know, so very unique cases. Because the problem with cadavers is that, you drill it once and that's it. And every cadaver is slightly different. So what you can achieve with this is that everyone can start with the same, everyone can start with this malformation or dead malformation, and then you can move to cadavers. We don't want to be, you know, a competitor to cadaver because cadaver is still...

the real thing, right? But having something where you can practice a bit more, we think would be really good.

Robin (11:14)
So the models you made, were you the ones designing it? You printed it, assume, and then you approached the EMTs or the institutes yourself? Or was it kind of the other way around where they approached you with very specific designs and needs and then you fulfill those?

Filip (11:30)
Yeah, as we were working on one of the other projects for medical device companies, one of the directors of that device company asked us, if we happen to have these type of models. And we started to, know, questioning, okay, if there is a market, if there is a need, we talked to our friends and colleagues here in Cambridge, because there is a fairly decent ENT community. So we were talking with them, seeing what could work or what couldn't work. And we started prototyping.

we started prototyping to see if it's even manageable to have a plastic or 3D printed model to behave as a bone. Currently, we are at iteration 44 or something like that, and there is always space for improvement. Basically, what we're trying to do right now is that we are part of a cohort in Birmingham where we tested out the bones with the ANTs there.

We tested it twice here in Cambridge. We are also running tests in Czechia and Prague. So we're trying to get as much feedback in regarding these bones and optimize the products, you know, where is the market and how we can make the best possible product basically based on the feedback from real surgeons and clinicians.

Hemdeep (12:45)
It would seem like as if this bone material that you have could be used pretty much in a wide range of applications. It could be used for the temporal, it could be used for any, you can change the density of the porosity of the bone material to mimic pretty much anything. So it could be used by any surgeon or any doctor throughout their studies.

Filip (13:05)
Yes, yeah, I think that we also started with one of the hardest tasks because the temporal bone is so unique and the anatomical structures are so small. So it's, yeah, I think we started with the hardest one, but we'll be fine in that, you know.

Hemdeep (13:20)
No, no, there would be no fun in that. You're right. Through your journey, you have probably learned a significant amount about yourself, about how to work in a collaboration in terms of identifying goals and then sort of understanding that the issues that you run into, whether they be problems in order to procure funding or to get adequate the right tools or

Right? Anything. What kind of things have you learned throughout this entire process?

Iwan (13:51)
Yeah, some of the lessons that I really kind of appreciated are the importance of validation and the importance of getting lots of different approaches to the same problem effectively, because no one model will solve every problem. So being able to use the, let's say, cadaveric models or computational models and use like kind of same analog in different ways.

has been really important and actually getting the validation as Philip was saying, and doing that independently to show like, okay, this is actually really replicating the same thing. Cause it's very easy to make a model, but how do you actually quantify it? It's true to the real anatomy that you can trust the decisions that you're making from that model.

because there's always limitations and understanding those kind of limitations is really key. And there's also being a little bit creative with the design aspect, I think is super important. So understanding of like, okay, well, this is limitation of this technology versus that technology, or how can we split this model into different component parts, have a little bit of more manual steps if needed, or go into different kind of multi-material.

kind of printers or different kinds of things when we win, need to. Because there's also considerations even with, let's say the drillable models that Philip just mentioned of cost. So it's not just making the best model, but especially for clinical schools where you need to have this as a consumable item, you need to think about how do you make it a cost effective and at a reproducible level where.

potentially you'd need to kind of tens or even hundreds down the line of these models. So how do you make it? You can make one very, very, very, very nice model. mean, the most effort is going into making that one model, but then you also need to think about how do you make that reproducible and even just making the post-processing very easy to do, easy to perform.

Robin (15:45)
So do you provide that entire workflow, like from the design, teaching them how to do the 3D printing side, the post-processing? Is that what part of Costa does?

Filip (15:54)
Not really. So what we do is to create the final product. we have basically two branches. One is doing the consulting work. So how that works, they usually, the medical device company or any company approaches and asks us if we can create something. And then we put our heads together and we try to think about how this could be done and if it's actually possible. That's one way we work. The other way we work is that on the background on

creating these bespoke models for companies, basically the consulting work. We create our own products. We do our R &D on things which we think could be interesting and could have a market. So then basically in the evenings we start running, you know, lots of experiments. And I wanted to follow up on what Hemdeep asked about any recommendations. I think like a practical example is that

You know, start with something small and very easy, you know, like not complicated structures and create it at least three times and then measure three times each and go this way. Because what happened many times, especially at the beginning, when we were starting with 3D printing, we sort of went too much in the future. created like this incredible difficult model. We printed it. It worked perfectly, but we were at

unable to replicate it. Every time when we tried to print it again or replicate the same thing again, it was basically impossible. There was always a problem. And then going backwards, it's significantly more difficult than going forward from something small. So I think it's a very casual sort of thought that you should work with a small thing and build up on that.

But in our case, yeah, that's 100 % true.

Iwan (17:53)
maybe one thing that which goes into kind of an approach we've also now taken with.

cost size, understanding like the ecosystem as a whole, because as well as working with medical device companies, we work more directly with clinicians as well as like, let's say medical schools and all of those bring different perspectives, but also different value in terms of understanding how you validate the models with directly with clinicians. And you have different ideas coming from those. then also understanding practical considerations going through with.

device companies and getting access to really high quality scans or like materials that we can then scan through medical schools that then can be used made more available because these things are hard to come by. You don't get cadaveric specimens of ears everywhere, right? So being able to democratize access to these kinds of models is really important to us. And we've even had lots of serendipity along the way, like just seeing

with a pediatrician, saw one of our ear models at a conference. It was actually not in Cambridge. They happened to also be around Cambridge and they had a very interesting technology looking at making effectively a microphone, which would feed into some bone conduction headphones for kids with recurrent ear infections. And someone actually had started calling their microphone, one of the nine year old girls actually.

called the microphone their extendable ear, because it's something from Harry Potter. The, the, some of the Weasley twins have an ear on a string basically. So they asked, can you make our microphone into an extendable ear? So we designed like the, the ear to be like a sleeve around the microphone. And then she conducted a whole study with the kids to see how they react and see like how they engage with the.

So there's a of really fun things which actually like helps like even though it seems kind of silly it's actually really informative to help kids to engage with it and she got another grant actually to then take those ears to Malawi as well. So she does a lot of work there. So there's now about 30 of our ears now in Malawi being used with these microphones.

Which is, yeah, quite cool. So was a very sweet story. Seeing the photos from it was amazing. Where especially like this small considerations, like they wanted to have them pierced. So then there were a deaf community in Malawi making earrings. So they can put the earrings on the bottles with the microphones. Yeah. It was a really touching story.

Hemdeep (20:26)
So you guys have had a very interesting five years. It seems that it went from not having really much idea of exactly what cochlear implants are about to now ⁓ running a company that is providing significant and interesting solutions that is cutting edge and you're reaching out to an extremely wide base of users and patients and

It's an amazing story to hear. think one of the mission statements that we have at the office is the fact that, you know, we see 3D printing as this ability for researchers and users to accelerate, you know, the idea to iterate, to design, to retest and so on and so forth. But eventually all arrows lead to one thing and that is to reach out to a larger audience, either through a commercial venture, through engaging with techniques. It looks like as if you have done it right and...

You've been very successful. I really do congratulate you for that.

Filip (21:25)
Thank you. Thank you for the guide.

Iwan (21:27)
That's excellent. Yeah. And I think so that comes back to like the way, all the way when we started the kind of looking at that 3D printer that we mentioned, we actually just got the Cochlear kind of models and sent it to a bunch of different companies and seeing like, okay, who can print it the best basically. So, and then from there, it's just getting the hits in, right? Like doing so many different iterations that you learn like failure and then you just improve 1 % each time.

Hemdeep (21:55)
Well, this was a fantastic inaugural podcast. I'm so happy that we had a chance to talk. And I think when you had presented to me exactly what you guys you were doing, it was amazing then and now to hear what you've done since. It's fantastic to hear. And I do thank you for taking the time out today to sit down with us. I think it was extremely meaningful for me and

and to Robin and I know our team is very appreciative and I'm sure that the audience that will hear this will also appreciate the work that you guys have done. Thank you.

Filip (22:29)
Thank you. Thank you for having us.

Iwan (22:31)
Yeah, thanks a lot for the opportunity and it's been great to have the support of Caddoworks because it one of the first printers we really had like a hands-on experience with. start, so it's like really kind of the start of this journey which has gone in lot of different wild directions as well. So it's been really good.

Hemdeep (22:47)
Amazing.

Robin (22:49)
And with that, that's a wrap on our three-part series of Iowan and Phillip from the University of Cambridge here on Big Ideas at Microscale. Over the past few episodes, we've explored their pioneering work in 3D printing and cochlear implants, from the creation of high-precision models to their efforts on improving implantation techniques by testing insertion forces.

Hemdeep (23:13)
We also dove into their company COSA Limited and how they're revolutionizing surgical training with realistic anatomical models. The impact of their work both in lab and through CASA is setting the stage for something truly exciting. A big thank you to Ivan and Philippe for sharing their expertise, their passion and their incredible story with us.

Robin (23:34)
If you enjoyed this series, don't forget to check out their ongoing work on their website at cosaltd.com. That's www.cosaltd.com. And then stay tuned for more exciting conversations in future episodes. For example, our next guest to be featured on the podcast is Jack Posh, a techno biologist working at the Aquatic Germplasm and Genetic Resource Center in

Louisiana State University. Here's a little sample of the episode.

Jack (24:22)
people go to the beach and like sunbathe or go swim in the water. But I had a seine net, seine net is, you know, maybe we can put the picture on the screen, but it's a giant net. I'd say 10 feet long by four feet high. Again, there's posts on each side. The bottom half of the seine net, there's some and then the top half, there are some floats. And so I brought a seine net to beach with me as like a seven year old.

And if you don't know what a second edit is, you post about. So they're pretty big. 

Robin (24:51)
Did you yourself bring it or was it like a family member had it and they like introduced it to you or was that just like you yourself?

Jack (24:59)
I think

I brought it.

We go into like the little store near the beachfront and I

saw one and I wanted to try it out. So my parents bought it for me, but I was the one who would bring it here. Whoever would come out in the surf with me, my grandpa, my grandma, my parents, whoever would drag the net to the surf with me, come drag net to the jellyfish and all sorts of things. So I think that's the main drivers why I'm so curious about the outside. But it's more than just an aha moment, I'd say.

every year. I would convince my co-

and we'd catch like fish.

What are they?

Robin (25:35)
Thanks for tuning in to Big Ideas of Microscale. If you enjoyed the episode, make sure to follow us and stay up to date. You can listen on Apple Podcasts and Spotify, or watch the full video on YouTube. You can also follow us for more updates and behind the scenes content on LinkedIn, Instagram, Blue Sky, and X. We're Cadworx3D across the board. That's spelled C-A-D-W-O-R-K-S-3D.

show notes, paper references, and bonus resources on today's topic, visit our website, catworks3d.com. That's spelled C-A-D-W-O-R-K-S 3D dot com.

Hemdeep (26:19)
Thanks for tuning in and as always, stay curious, keep exploring and never stop asking the big questions that are shaping our world. Whether you're in a lab, on the go or just curious about the future of technology, join us as we continue to dive into big ideas at Microscale.