MONDAY, 29 JUNE 2020
Below is a transcript of the latest episode of the BlueSci Podcast, which we have provided to help our podcast reach a wider audience. The script has been transcribed using an audio-to-text converter (otter.ai) and edited by our hosts. If you spot any errors please do let us know! Please do subscribe, follow, like, and review our podcast to show us your support and give us feedback, whether you listen or read the episode.
You can find the podcast on:Printing the future, with Tanyaradzwa Mangoma
AnchorFM : anchor.fm/bluesci-podcast
Apple Podcasts: podcasts.apple.com/podcast/id1358079746
Google Podcasts: podcasts.google.com?feed=aHR0cHM6Ly9hbmNob3IuZm0vcy8zZWRmMTdjL3BvZGNhc3QvcnNz
AnchorFM : anchor.fm/bluesci-podcast
Apple Podcasts: podcasts.apple.com/podcast/id1358079746
Google Podcasts: podcasts.google.com?feed=aHR0cHM6Ly9hbmNob3IuZm0vcy8zZWRmMTdjL3BvZGNhc3QvcnNz
Released 29th June, 2020
BlueSci Podcast 0:05
Welcome to the Bluesci Podcast, brought to you by Cambridge University Science Magazine. I'm Ruby and I'm Simone. Every two weeks we speak to local researchers, university staff and students and anyone who works in science to learn about their research and activities, hear about the work that they do, and uncover what goes on behind the scenes. If you want to get in touch with a question, suggestion or just want to be featured on the podcast, just drop us a tweet. Our handle is @bluescipod and you can also email us at firstname.lastname@example.org
Today, our guest is Tanyaradzwa Mangoma who is a PhD candidate in the Fluids and Advanced Manufacturing group at the Institute of Manufacturing and works in collaboration with the bioelectronics group in the Department of Engineering. Her research is focused on fabricating proof of concept additively manufactured neuromorphic and neural network devices based on organic electrochemical transistors. We spoke to her about her research, what it's like working in such a fast paced interdisciplinary field and what the future could hold for personalised by electronic medicine.
The BlueSci Podcast is sponsored by Greiner Bio-One who supply laboratory diagnostic and medical products to research institutions, higher education, the NHS and others across the UK. For details of the full product range, visit www.gbo.com.
Welcome to the BlueSci podcast Tanya. You're in the field of fluids and advanced manufacturing research. So can you tell us what is your research on specifically?
Um, so my research tries to use additive manufacturing, commonly like the public uses 3d printing, to print our electronic devices that are called neuromorphics. And neuromorphics are basic devices that work like the brain and the neuromorphics that I use, they're based on electrochemical transistors, organic electrochemical transistors actually. It all comes together that 3d print this component that works like a transistor but also like a brain. And then I use all the manufacturing knowledge that I have from the fluids and manufacturing team to kind of see the best way to actually print the devices.
So does that require doing the device development as well? Or just learning? Or like looking at the processes that are used a manufacturer? Or do you have to look at the research on the actual bioelectronic side as well?
So you have to do like the whole thing. So it's, well, I have to design my own devices. So like part of my research is actually designing a device. So the different devices with different things, some devices just like the one kind of thing that works Others have like 20 other things attached to them or like they work as a group. So it's more about designing them and then after you desin them, it's about how they would actually be manufactured. And then each manufacturing step you have to like make sure it actually works. And it actually qualifies that part. And at the same time, you need to know the electronics bit because like, without knowing how the device works and the electronics you can't actually manufacture it. It's like it's like all like together is Yeah, it's like the whole spread Yes, know how it works, and how to manufacture it and whatnot. Yeah, the whole thing
So it sounds quite, quite a challenge then to go from the sort of basic science of it and a sort of prototype device all the way looking through to how it how it can be manufactured. And, and that process. It sounds quite expansive. So you mentioned the additive manufacturing, and like you said, the most common example of that is 3d printers. So apart from that, what are sort of the other big products that come out of additive manufacturing? are there different materials that you can use? Or what sort of things can you make?
So like additive manufacturing is a blanket term for a bunch of different techniques. So any material you can think about - metal, ceramics, polymers, you know, you can use them using a certain type of additive manufacturing technique. So products can range from car parts, space parts, that's what people usually think about in additive manufacturing, like, you know, in industry, and they can go down to like our plastics things, like toys and stuff like that. So it's a very big range of things. And people are even doing up biological tissues, for example, they do via 3d bioprinting. So it's very, very varied. So any material you can think of, if it needs to be manufactured, and you can like put it on top of each other using layers. Then you can in theory, 3d print it. So just think of a product and we have it for you. Mostly though I do focus on polymer based materials, because that's what I stick to.
So within additive manufacturing knowledge, there's all these different types of processes in terms of neuromorphic devices specifically, I guess, what are the like key challenges in that field? Is it the actual development of how you're making like the design? Or are there also challenges to the actual printing itself, because you're dealing with different kinds of materials?
So I would say the additive manufacturing techniques I use are something called fused deposition modelling. And I also use something called geo-extrusion and the third one is inkjet printing. In all those three fields, there are individual challenges. For example, in fused deposition modelling, FDM, the accuracy is horrible. The precision is horrible, for the technique in general, and, for instance, FDM is the 3d printers that you see people using plastic to make toys with. So the material development for that if you want to have something that's electronic and conducting, it's not developed, to actually makes something that is rigorous for science. So like the challenges I do face, the challenges I'm like, Okay, if this going to work in like an actual setting, who needs to have a material that is specifically made for this application. And at the same time with Inkjet printing, it's been-Oh, by the way, inkjet printing is similar to the printing that we do on paper, like in offices, so like that type of printing - Um, they've developed it for newspapers, but they developed for electronics, but it's not rigorous enough to be like, okay, we're going to do this to form your neuromorphic devices, because you have to remember, your neuromorphic devices are supposed to work at a molecular level, like they're like literally, like, trying to act like the brain, using molecules. Like at that level of precision, things need to be perfect. To get that, we have to go back to like, the materials being made, the equipment, the accuracy, how to integrate the two kinds of technologies together. So these are all challenges, like you know, I think about just you know, in general, it's a it's a lot of difficulties throughout the field. I mean, it's far from perfect,honestly.
You look specifically at neuromorphic devices and your research is centred around that. Could you explain exactly what what your project within that is and the type of devices that would typically come out of a project like yours?
So with neuromorphic devices. Like I said before, they try and work like the brain. And in particular, the ones I work with, they do this really cool thing, I think, action potentials, basically, so they spike. So they spike like action potentials in the brain. And they actually retain some of the memory and that is like the basis of learning in the brain. So in the synaptic level. And what my devices try to do is to mimic that but not in a device that is made from like silicon technology or lithography. But using additive manufacturing. And the point from that part is well if we can do a device or additive manufacturing device that works like the brain for applications that have been identified in OECTs, which include drug delivery biosensing and brain machine interfaces, they need to be specific to that person. Like, if I'm going to the hospital and I'm an injured specific way, I'm gonna need a specific device to kind of acknowledge this specific thing. And my, my topic is just being like, okay, so we could additively manufacture this because to do that you just need to make a CAD model that is different. So mine is more of a visibility check on if it's possible to actually make those devices and after that, how complicated can we make this devices? This was another part that goes hand in hand with neuromorphic devices is neural networks. So kind of like you know, in the brain when you have a couple of synapses together you make a neural network. So could we make a neural network from additive manufacturing and how complicated can we get? can we use it as a technique to actually study what is happening in the brain, or study, how people are actually capturing data in machine learning for instance. So that is like the range of what my PhD does, trying to make the proof of concept devices which are basically: Okay, can we make this so that if it was going to be used in an actual setting, individual people can have like customised, personalised bio-electronic medications? And then at the other end, we have, could you make it super complicated so that we could actually use this as like an actual, like machine learning or in this case, we'll call this deep learning architecture. Like that is the extent of my PhD,
I mean, it's sounds super like, like you're spanning so many different research fields and so many different questions very, like interdisciplinary, I guess, what was your background coming into what made you soe interested in this specific problem?
Basically, I knew nothing. So because I was part of like, CDT like a centre of doctoral training. I came with, and I could just do whatever PhD I wanted. I didn't know how to do additive manufacturing or what it was. I didn't know how to do any modelling or CAD drawing. I didn't know anything about neuromorphic devices, but I really wanted to do it. So I was like, Oh, you know, this sounds interesting. So like, I just asked the supervisor, you know, could I actually do something called 4d printing? And he was like, oh, there's another guy who's like printing brain stem stuff. Do you wanna do that? It's like, yeah, sure, you can do that. So like, that was my background, there was nothing concrete that like helped me. My background didn't have anything to do with this, but it was just nice having this field, that was crazy. And really, really massive to just play around with it. Yeah.
That's amazing, because it just shows how flexible the whole PhD processes and yeah, if you if you find an interest in an area you wouldn't expect then it just shows you, you can do that. It's really cool. Definitely. And so it seems like this field in within the additive manufacturing context is fairly young. Would you say that's true? And so what is the state of the field at the moment, like what what, what's kind of generally known and you know, because I guess when when you think about the use of bio electronics, see it my mind instantly goes to sort of crazy implants and like, you know, these different movies and stuff. Yeah, yeah. And and so I guess I guess how far are we from something like that, especially in this field where it seems, seems quite, quite new?
In the groups I'm in one of my groups is in bio electronics, so I'm shared between two groups. And in that group, they literally just focus on devices that are made on the science I tried to print. And they put it in like animals and cadavers. And they do amazing things, they've done it for epilepsy, where they put like actual devices in the brain, and then they stop seizures of epilepsy. They've done something called Deep Brain Stimulation, where someone I forgot exactly what they have, but they can't stop shaking their hands and having tremors and in the brain, they kind of stimulate it and they stop and they can actually have the mobility back. And so this sort of bio-electronics in itself is being developed very well and very rapidly and it is based on this organic electrochemical transistor materials and some of these really unique semiconducting polymer materials. And then when it comes to additive manufacturing on the other hand, it's now starting to be used to be used for electronics. So people are starting to just do general electronics, using additive manufacturing, they make electrical contacts that can recover by time, they are starting to develop inkjet devices on their own, that actually can be used transistors. But bringing the two fields together is very, very novel, like not many people have actually made neuromorphic devices in general, not many people make them that are additive manufacturing. So it's a really unique like one of those like really kind of cutting edge. And even if you were to find a device that is being printed, using which is neuromorphic and is also used in some type of manufacturing, it's highly unlikely that they use two types of additive manufacturing together. So I use two, deposition modelling and InkJet printing together. And I'm very one of those like, really kind of like, you know, not many people are doing it. But if it's done, you know, it's very useful. Yeah, it's interesting. But there are developments being happening around the field in general.
In terms of applying kind of additive manufacturing techniques to bio electronics, often when people think about, like brain enhancing or like, you know, kind of performance enhancing devices that can be used in the future, we'd often come up with this, like, kind of ethical question of like, well, who will have access to these very expensive things? And, you know, how can we make sure that, you know, they're not using like a malignant way and things like that. So I guess, on the other hand, as manufacturing is a very, like very easily scaled way of producing things, do you think if we were able to use that as manufacturing for bio electronics, and just in general, I guess, do you think it's kind of like an equaliser in that sense to like making more cheap devices and stuff?
So to be honest,especially for the one I use like the printers I use literally go for 300 pounds and the material goes for like five pounds, so really accessible. So the one thing that I really like about this kind of a concept of using additive manufacturing and inkjet printing, which again, I said, I can take the printers in your little printers, if you put on paper, they're really accessible. Many people could actually just buy them and use them to actually make products that work. And I mentioned earlier on, you could use them for biosensors. And if you could actually sense for example, like or detect a certain type of disease or something, using the additively manufactured, easy to make devices. It's easier, let's say pick a country that is far far away. And you just send the 3d printer and some material just like print as you like whatever thing you want to do for detection. That makes the process extra easy. You don't need a cleanroom you don't need expensive, like, you know, facilities, you just need that thing. But then at the other end, if you make it very, very easy for people to make, and then we introduce this whole idea of bio electronics in it, where you can actually then have people data and like health monitoring, yeah ethics are questioned. And you know what? Yeah. ethics are question because like, if I can print and like take data from people and what they do then yeah, it's a tricky place.
Yeah.Yeah, that that is that that sounds like quite a conundrum, really, because you simultaneously want people to have access to this type of sort of, you know, product, but equally, you don't want people to misuse it. And so that's really, really hard. And, and do you think that within this field, there is a push push towards looking at these sort of things like, and so for instance, if you were to design something that that would kind of be released, so in a way that people could print it and use it themselves like, would data privacy and, you know, agreements have to be brought in and legalities and things like that. Is it something that you have to collaborate with in terms of the legal side of things?
I mean, it actually depends because it could work in one really good way that you know, when you have a device, and because it kind of senses, it, it also delivers. So you have a device that is put on your arm. And it just senses and delivers, and no one can get your data. So it's kind of secure in that way. So there won't be an agreement. But then it's like, if somebody goes like, oh, but for me to do it, I have to put a chip in it to record every data point that has been produced, then we'll have some legality issues that at that point, so like, it's basically how good companies are going to be if they actually take it up. If companies are like, we're just doing this to save humanity, then they wouldn't put a chip inside to detect everything that is happening, because there is like really secure but then if there's like really, really companies like data? Which a lot them are, then again? Yeah, another one of those, you know?
Yeah, I guess I guess it's hard isn't it? Because I guess, in some countries in the world if if it if it were a way of monitoring a certain bodily state say like, blood glucose or or anything like that, and it was a country where you have to pay quite high health insurance premiums anyway. And somehow that health insurance companies got it. And I guess there can be a lot of like, knock on effects.
I think there's always those kind of questions when it comes to like medical ethics, everyone knows about your health records and so on, I guess I would just make it harder to to know because it wouldn't be a piece of paper, it would be like some data in a chip that you as a normal patient, you might not even know is being collected. So yeah, that's, that's really interesting. Um, I was going to ask, in terms of the actual way that you're going back to kind of focusing on your work, how does that kind of work day to day right? Cause like I'm just imagining these like machines that print things. But then do you how do you end up like testing the devices to see if they work? Do you have to, like how do you come up with all these new ways of, of seeing, like testing the viability of the proof of concept devices that you make? How does that kind of what's the day in the life? You know?
Oh, there is no day which is one thing I absolutely love. I've yet to be bored with my PhD because there's no day that is like the same. So like, some days I am designing, some days I'm in the like, electronic like lab, some days I'm like cutting the device in half trying to figure out why it's not working. So each day depending on which point I am in, like making a device is different from one another. So I do totally different things. Sometimes I'm manufacturing sometimes I'm like reading theories it's very dependent on the day and what I do, but for the most time, I am sitting down reading to understand what the hell I'm doing because like, it's either like doing.. Each part I'm doing for my device. It's always gonna be something I don't know. So I'm like reading about it, and then I do it. So it's always different. But if I'm going to say what I say my most time doing, it's reading to trying figure out what is happening, because everything goes wrong on this. But yeah, there's no specific day.
Yeah, and it sounds like, as we mentioned before, like it, it sounds so interdisciplinary. So with respect to that, do you have quite a lot of collaboration with different people? So for instance, if it's neuromorphic devices, are you in contact with neuroscientists or bio people? And if so, how do you find the interactions with them? Is it? Is it easy or can it be a little bit difficult to try and for each side to try and get their point across?
So I'm already in two different labs and my first lab, they do like a lot of manufacturing. So if I don't know how to manufacture I just go to one of the people who's manufacturing. The second lab, they do it bioelectronics. So if I want to take us something about, you know, how does this work in the electrical, electrical engineering people, it's all in there. I just asked him as well, so I am really lucky and fortunate that the two core groups that I do need, the two core parts are like bio electronics and additive manufacturing that I do need. There are two groups that are just there for me to access. And then when I don't understand something, these are professors like in other parts of the world, just so I go like, so how does your device works? Because it doesn't make sense using any theory. And they they're really responsive. So it's a good yeah. And the interactions themselves, I find them quite challenging because they speak different languages. If I look at people who are doing manufacturing, and they I'm talking about like, electrical engineering, where they do totally different things and it's like, translating that into understanding and being the middleman to all that science is quite interesting, because you get like a big group, but then I'm like, oh! But yeah, it's like I have the access to them, but then it's very difficult kind of like get it into one language that everyone understands.
Yeah, no, I mean, that sounds quite challenging in general, whenever you're like an interdisciplinary field, but then at the same time, there can be some benefits to seeing the same problem from so many different angles as well. Do you think that like the, that maybe, perhaps, if we start to have those conversations more like in terms of people that are maybe more on the more basic research and then people are further down the long line of like, the more engineering aspects of that, do you think that could maybe lead to a lot of problems being solved faster? I guess, if people weren't so like, in their little bubbles?
Definitely, I mean, honestly, like, I take myself as an example, like, you know, like, there are computer people who are doing computer stuff and there's like, you know, neuroscience people doing their neuroscience stuff. And then there's like electrical engineers, like, I'm like, if they all came together, like you could have like, deep learning like architectures, you know, just like that. Just throw money at it. But again, it's very difficult because in those specific fields like if I was to take for example, in neuroscience, there is a whole lot more to be like understood that is way too deep for like an interdisciplinary field. So I do appreciate people in their fields who are deep into it. But we do need more people doing interdisciplinary stuff because of those collaborations comes really nice things like really amazing. Like, honestly, if you look at, especially in the bio electronics group that I am in right now, people couple to very different fields together. And out of it comes really amazing work that you could never think of alone a specific field, you do need to have that exposure to other things. So I appreciate fields are big and they have deep stuff to be learned but collaboration and interdisciplinary fields. Very important.
Yeah, no, I guess you always get something unexpected happening out of it because it required that those two ingredients to actually make it happen. Yes, interesting.
Yeah, I bet. I bet it keeps it very relevant as well. Because, yeah, like, like you said, sometimes, if you stick in your field, you go so deep, and you forget why you're there. Whereas if there's a common goal across disciplines, then perhaps it's slightly more relevant to what's needed in science right now in general. And, and so, with respect to this, obviously, you came into your PhD with a very flexible mindset in as much as what you wanted to study and, and you were happy to sort of go into a field that you perhaps weren't as familiar with. And what what what are your hopes for the future? Would you like to stay in academia? Would you like to go into industry? Would you want to go for something completely different? Is it something you thought about or is it just, you know, get the PhD and then I'll think about it type situation.
I thought about it a lot after like, the first couple of months of my PhD because I was like, What am I doing? Like, really? You tell people you're doing this and like you just throwing you know, buzzwords around; additive manufacturing, neuromorphic... you know what I mean? So like, what exactly am I doing? I did my research in who's working in this right now who's doing neuromorphic computing who actually wants to make this stuff. And I found a couple of really good companies and like I fell in love with IBM, like straight away. So like, I definitely want to continue what I'm doing. And I definitely want to develop the field of additive manufacturing and for bio electronics, especially on neuromorphic devices. And I'm hoping when I graduate, I actually get a role in like a big company. But it just develops because I really see the potential but I understand there's a lot of work that needs to be done to like bring it all together, because starting from the materials themselves, they need to be better, the equipment itself needs to be better. There is so much work to be done. But then after it's done, the impact is amazing. So I really would like to stay in the field and actually continue doing exactly this
And in terms of the kind of hopes and dreams for the field, I guess what are the kind of main not not main targets but what can we expect from bio electronics in the future? I guess what could, where like if people if people finish listening to this podcast and someone asked them what you know, what was the podcast the podcast this episode listen to what was it about and they go oh yes neuromorphic devices like what can we expect from neuromorphic devices in the future?
I hope in the future when you go to your doctor, and you know, you realise you have this specific pain and it requires this much paracetamol, they'd just print off a device for you put it on your arm, for instance, tailored to you. And it will just dose you with however much paracetamol you need and detect how much has been given, or your response, over time. And that will like cut, it will be like one of those like groundbreaking things, basically, in medicine. That is like my hopes and dreams, but who knows. It'd be pretty nice if that happened.
It sounds like such an exciting field and emerging field. And so thank you so much for filling in filling us in about it. I mean, I for one had very limited knowledge of this area. So it's been really, really interesting to chat with you about it. And I really, really hope all of your PhD work can get going again soon. Off the back of a lockdown.
Thank you so much. Thank you for inviting me really nice. And every PhD likes to talk about their PhD.
We really hope you enjoyed this week's episode with Tanya talking all about additive manufacturing and neuromorphic devices. If you enjoyed this episode and want to listen to some more, you can find this on anywhere you can get podcasts and give us a like, subscribe to us and then you can have a look through our back catalogue. Listen to all our old episodes and also stay notified whenever we have a new episode. You could also contact us on Twitter @bluescipod, and you can also email us at email@example.com. That's it for this week and we'll see you next time.