Transcript June 2007
Gareth Mitchell: Welcome to the official Podcast of Imperial College London. Oh, and we're Europe's leading science university. Just thought I'd mention that. And I'm Gareth Mitchell, hello. I'm presenter of the BBC's Digital Planet and also a lecturer here in our Science Communication Group. This month, plans to control red deer populations in Scotland. Good conservation or crying wolf?
Tim Coulson: Unsurprisingly, if you introduce a predator we will end up over the course of time seeing changes in deer densities, and deer numbers would go down quite a lot. That's being called for by quite a few conservation organisations.
GM: And from wolves to the ward as one of Imperial's junior doctors reinvents the tourniquet.
Christian Fellowes: The pressure in the venous system is obviously quite low so you don't need a huge amount of force for it. Here on your arm we can see that the veins themselves are swollen up now, perfect for taking blood.
GM: But why exactly does the world need a new kind of tourniquet? Find out about that and other headlines from around the world's ninth best university no less here on the official podcast of Imperial College London.
Russell Cowburn's anti-counterfeiting technology
Well, to start with, could barcodes be a thing of the past? Well, that's a good question for Russell Cowburn. He's Professor of Nanotechnology in the Department of Physics here at Imperial. And he's one of a team behind a new technology called Laser Surface Authentication. Sounds intriguing. Tell me more Russell.
Russell Cowburn: LSA, Laser Surface Authentication. It's like biometrics. So if you think of what you know about human biometrics. Things like fingerprints or iris scans. Things that allow us to identify humans. Laser Surface Authentication is the equivalent of that but for things that aren't human. For things like sheets of paper, boxes of pharmaceuticals, anything really. It's a way of finding a unique identity code based on the intrinsic properties of different items.
GM: And that's quite an intriguing thought because for humans you can see how there are things that mark us out from other humans, like our fingerprints or our retina, whatever. For instance, you've got some pieces of paper on the desk here and each of these three pieces of paper look pretty much identical to me.
RC: That's right, and certainly to the human eye they're indistinguishable. If you mix them up I wouldn't know which one was which. But if we look at them with very, very high power vision, if we could see down to the nanoscale and actually look at the individual fibres and even smaller, then what we'd find actually is that they are as different as you and I are different. That on a small enough scale they have features which are completely unique to them.
GM: And at that incredibly microscopic level then one seemingly identical sheet of paper is different from another?
RC: That's right. There are features in all surfaces which are random, and the exact detail of those features are set as the item is manufactured. So in the case of paper it's the way the fibres settle as the paper is formed from the pulp. But it's not just paper. It's true of plastic. As a plastic card sets from the melt then again the surface ends up having microscopic random properties. And in fact as a physicist, if you just look at anything, as long as it doesn't appear to be a perfect mirror, then from a physics' point of view the reason that it's not a perfect mirror is because there are these random features on the surface. And what we've got is a technology that allows you to access those and to use them for security and authentication and anti-counterfeiting.
GM: It's as if these objects have their own inbuilt barcodes accept these are attributes that are already in these bits and pieces, whether it's paper or plastic or whatever?
RC: That's right. So just as nobody gives you your fingerprint or nobody gives you your iris scan, we haven't had to do anything to these items. It's the way nature has given them to us. They're self-tagged, if you like, already. And that's what's exciting about this, that we can protect things from counterfeiting without having to modify the item in anyway.
GM: And when I heard that you had a technology for looking at these incredibly small details on bits and pieces of paper, or whatever they are, I assumed that you'd lead me into a room with some kind of scanning tunnelling electron microscope or some optical bench that was going to take up a whole lab. In fact the equipment that you have here sitting on your table, we're not even in the lab, we're in your office, and you have a laptop set up and plugged into it is a box, which is actually a bit smaller than the kind of set-top box you might have on top of your TV. And this is where, if you like, the clever stuff is, in this box here?
RC: That's right. I mean we could have taken you to a lab and shown you an enormous microscope. And indeed we've got lots of images of what materials look like on the microscopic and the nanoscale. But that wouldn't really be a technology. You don't want to have to put big expensive microscopes at passport control or at point of sale in pharmaceutical outlets. The inventive step that we've come up with is finding a little laser scattering method which can be built into a very small portable unit, relatively low manufacturing cost, but which gets us all the same information as if we had a giant microscope there.
GM: I'm assuming then that you can use this device plugged into your laptop to tell one of these pieces of paper apart from the other?
RC: Yeah, shall we do that?
GM: Let's give it a go.
RC: Okay. So we've got three sheets of identical office paper here.
GM: It's just normal A4 paper, isn't it, white paper?
RC: Absolutely. They're unmarked so there's no way of knowing which one is which. Two of them have been shown to the system before and we've recorded their naturally occurring fingerprints on a database. The third one has never been shown to the system before. And in terms of using this as an anti-counterfeiting technology the two that have been shown to the system before, they are genuine things because they've been registered, if you like. Whereas the one that hasn't been shown to it before is the counterfeit because it's never been seen by the brand owner. So let's take one of the ones that should be known. And I'm just going to lay it on top of the scanner here and it'll take about one second to scan and you'll see a laser beam whiz across the surface there. So here we go.
GM: Yes, I saw the laser beam. It's a bit like those scanners that you have in a supermarket that reads a barcode.
RC: Yeah, that's right. It's the same type of laser but obviously the analysis is more complicated. And it's looking at a very different thing. It's looking at all the microscopic structure of the i tem. But as soon as the scan finished you'll see here on the computer screen it's c ome up with the na me, ‘sheet of paper 1', which is what we called this sheet when we registered it. Now, let's take the next one, which remember to you and me is indistinguishable. Let's see if the system confuses it or if it can tell them apart. And here we go, another scan. And you see that one says, ‘sheet of paper 2' this time. So it can tell the difference. But then as the real test let's take the counterfeit one, the third one, and you see it says, ‘item not recognised'. So it knows which ones it does know but it knows which things it doesn't know as well.
GM: That is absolutely astonishing. Three seemingly identically pieces of paper and you've just proven there in front of me using this desktop device that the computer can tell one from the other and spot the counterfeit one. And I have to ask at this point, how?
RC: It has to be said we are still amazed.
GM: And you invented it.
RC: Exactly. I'd like to say that we expected all this and we predicted it from the start but on a daily basis we are still gob-smacked by just how well it does work. The physics is relatively straightforward. If you shine a laser beam on to a surface, if that surface is flat to within the nearest atom or so, then the light bounces back the way it came and nothing scatters to the side. But as soon as you've got any sort of imperfection at all in that surface, and even just a few nanometres of imperfection, then that incoming light is scattered sideways. And we see this not just with lasers but with everyday objects. If you just hold up the sheet of paper it looks bright and it looks white. And what it's doing is it's directing light from the window there into your eye even though the paper is not aligned in a precise mirror like fashion. If this weren't a sheet of paper but were a mirror then I would only see the window if I got the angle just right. And that's the difference between what physicists call specula and diffuse reflection. Diffuse reflection is where the light goes everywhere. Specula is where it just bounces off from the direction that it came. So we've got little sensors that only look at the diffuse light and therefore that light is only due to the imperfections in the surface and so it allows us to focus in on the thing that we're interested in and to largely ignore all of the other things that are going on around there.
GM: When you scanned the bits of paper just now I noticed that the laser was really looking at an area, I'd say, towards the top left hand side of each sheet of paper. Does that mean that had you scanned the bottom right hand side, for instance, that it wouldn't have noticed it?
RC: That's right. And that's an important part of the technology. Because, for example, the way paper is made is on a large drum and it's then only at the last minute cut up into what we think of as A4. It actually starts its life a much larger single sheet. And so we don't want something that's only specific to a sheet. It actually has to be to even different places on that sheet. On the other hand that means you've got the technical challenge of using a technology like this of making sure that you're examining the right part of the item. And so we've got a number of strategies. You'll see on the scanner here we've got a little bracket that holds the paper and that guarantees that you're looking in the right place. But also we've got a little camera, a little web camera, built into that unit which can actually look for little logos or little markers on the surface to help it to find the right place to scan.
GM: And until I'd seen this demonstration about two minutes ago I'd have assumed that the only way you could tell one inanimate object aside from another at that kind of microscopic level would be if actually on the macroscopic visible level you had a barcode or some other kind of identifier. Here you don't need to modify the paper or stick anything to it or anything. The system just recognises it. And that I assume is what gives this a potential commercial edge?
RC: That's right. Although interestingly you might still want to keep your barcode. The system works particularly well if you combine this with one of the more traditional things like barcodes. There's a lot of interest in, for example, stopping the counterfeit of pharmaceuticals in putting a little unique barcode on every item. And the idea is when I get my box of pills home I can type the number into the Internet and it will tell me, yes, that's a genuine item. As long as you chose your numbers carefully the chances that a counterfeiter could think up a number which is correct are vanishingly small. Now, the way that system falls down is that all the bad guy has to do is go along to his local pharmacy and take a look at a number of a genuine item and then make lots of copies of it. Now, with LSA we can link the barcode to the actual physical item that it's attached to and stop people from making copies. If you combine this with barcodes it's a way of making an almost perfect security system. But it's to easy to read on the one hand but on the other hand it's impossible to copy.
GM: But how sure can you be? I don't know if you can measure it statistically about how secure this whole system is. Can you absolutely be sure that you can tell one piece of paper apart from a thousand pieces of paper, a million, ten million? Is it robust enough for that?
RC: That's a really important question because one of the ways to defeat a system like this is just to keep drawing random sheets until you accidentally stumble across one that shares the same signature as the thing you're trying to copy. So we've done a lot of work on the statistics of that. The probability that two sheets share the same signature is somewhere around one part in ten to the power of a hundred. So that's a billion, billion, billion, billion, billion, billion, billion, billionth. That means that unless you've actually got ten to the one hundred sheets available to you, you won't on average find one that shares the same code as something else. Now, ten to the hundred is actually more than there are atoms in the Universe so there isn't enough stuff out there to even do the experiment.
GM: Not enough sheets of paper, not in all the phone books of all the world?
RC: That's right. But it's a serious point. It means it is really unique and it's really discriminating. The technology is very able to tell things apart even if you've got billions of items that it could chose from.
GM: Even so, you are up against a whole load of new authentication technologies. I'm thinking especially RFID, you know, like these little radio tags that you can put into bits of shopping and so on so they can be scanned quite easily. People have a lot of choice now, don't they, when they want to use systems for authenticating products?
RC: They do and it's important to think really clearly about what are the threats that you're trying to stop. And I think RFID is a really interesting example because RFID is a brilliant technology. But it's not a security technology. It's a tracking technology. The only reason people think it's a security technology is that it's not that widely deployed at the moment and so it's fairly rare and that gives it a certain difficulty to counterfeiting. But as soon as we're in a world where everything comes with an RFID tag commensurate with that will be Internet sites where you can just dial up the number of the RFID chip that you want and you can receive a thousand through the post the next day. I mean, there are places like that that exist already today. And once that happens it's no longer an authentication or a guarantee it's just a convenient way of keeping track of all of the goods in a warehouse or getting them through a checkout quickly. So we think you need to combine technologies. LSA is great for securing things and is great for tracing things. Combine that with something like RFID and you've got a really top solution.
GM: And this isn't just bonkers blue sky stuff that's confined to the lab. You've now set up a spin out company to market this commercially haven't you?
RC: That's right. So we have a company based in central London called Ingenia Technology Limited. And Ingenia exists to really bring this science and this engineering into the real world and to actually get it on to the production line of brand owners and get it into the documents of people who have things to protect.
GM: And let's let you blow your own trumpet as well then. You've won an award?
RC: Yes. In April we won the Hermes Award which is part of the Hanover Fair in Germany. The Hanover Fair is the world's biggest engineering fair and there's a prize associated with some of the best technology on display at that prize. And we won that in conjunction with our partners Bayer Technology Services who are part of the pharmaceutical giant Bayer. And together we've brought a product to the market that people can use to protect their goods on a production line.
GM: Well, a great story to kick off this edition of the podcast. So Professor Cowburn, thanks very much indeed.
Headlines from around the College
Russell Cowburn there. In a moment wolves and tourniquets, though in separate items of course. But before that though this is the news' music so here are some quick headlines from around the College.
The long held theory over how the HIV virus depletes the body's immune system may be wrong. So says an Imperial Professor of mathematics. HIV attacks infection fighting cells in the body gradually knocking them out and hence weakening the immune system. The thing is the decline usually takes about ten years. So why is the process so slow? A mathematical model of how the virus destroys immune cells questions the so called runaway hypothesis. This popular theory says that HIV infected immune cells produce more HIV particles which in turn trigger more immune cells which then get infected leading to an avalanche of disease. Professor Jaroslav Stark says that his model should make scientists think again about how HIV works.
If you could understand what the process which is leading to this very slow decline is then you might have a hope of focusing your therapy and ultimately halting it.
And a new all-in-one energy generator fridge and cooker sounds like a good idea, with the emphasis on sounds. Imperial College mechanical engineers are working on thermoacoustic technology where sound waves are converted into heat and vice versa to design a refrigerator for people in the developing world. Thermoacoustic technology is still pretty cutting edge stuff but the idea is to adapt it so it can be easily manufactured and maintained to power cooling units much more efficiently than by conventional methods. Then it'll mean reliable, affordable fridges ideal for storing medicines and foods in some of the world's poorest countries.
Reintroducing wolves to Scotland
And you can keep up-to-date with the big stories from Imperial before the rest of the world reads about them in the newspaper by visiting our Press Office website. Just go to imperial.ac.uk/news and there you will find stories like this one. A tale of Imperial researchers, wolves and bird life in the Scottish Highlands. It's all about controlling numbers of red deer. In some areas populations of the deer are too high and they disrupt woodland habitats with adverse consequences for plants and other animals like many species of birds. Perhaps it's time to reintroduce a predator and so enter the wolves. This report from Alison Kirby.
Tim Coulson: We were interested in looking at what the consequences of reintroducing wolves to Scotland would be on the Scottish red deer population. So unsurprisingly if you introduce a predator we will end up over the course of time seeing changes in deer densities and deer numbers would go down quite a lot. That's being called for by quite a few conservation organisations. So, for example, an organisation called Trees for Life is very interested in reforesting parts of Scotland. By reforesting areas and by modifying the landscape you would also get birds like the capercaillie and other wildlife species coming back in greater numbers and their populations increasing. Obviously, if you have too many deer the deer tend to eat the trees and the seedlings so you don't get natural forest regeneration. You end up with a very specific ecosystem with lots of deer. It's costly to reduce deer. There are financial implications. So one route to reducing would be to reintroduce a predator. And our conclusions would be that the density of deer would reduce to about somewhere between a quarter and a third of what it is now. A good example of where that's actually occurred is a place in Canada called Algonquin National Park where they've had wolves return naturally about 50 years ago. And all the studies in that park reduced elk, which are the same species as our red deer. Elk numbers have declined and they've seen an increase in their numbers of many other species.
Alison Kirby: Hearing of this possible venture I decided to go along to the UK Wolf Conservation Trust in Beenham to get face-to-face with some wolves for myself.
John Dennis: So this is the alpha female European wolf. This is Lataya.
AK: She's lovely.
JD: She is. She's lovely. And these, the two black ones, or the dark ones, are sort of almost certainly Mackenzie River Valley wolves in as far as they have a tendency to throw darker wolves, and we think that's where their parents came from anyway. And he's a cross between a European wolf and a North American wolf.
AK: That was John Dennis, Senior Wolf Handler at the UK Wolf Conservation Trust.
JD: If you were to put two or three packs of wolves into Scotland, into the central Highlands, they would have very little impact on the deer population that's there at the moment. What you would have to do is; man interfered and did away with the wolf, the top predator, the apex predator. Deer were protected and deer have thrived and now they're at the point of eating themselves out of house and home. So man would have to intervene again and cull out a vast part of that population of deer so that the wolves could then go in and maintain that balance of nature.
AK: Although methods to control red deer populations in Scotland are currently in place there are significant problems, as Dr Tim Coulson describes.
TC: So at the moment deer throughout Scotland, the culls throughout Scotland, are organised by an organisation called the Deer Commission for Scotland, and they have specific management objectives. What they do is they go to estate owners who tend to own the land where the deer occur and they will ask or tell the estate owners that they need to kill a certain number of deer on their land, which they do through shooting. However, the income obtained through shooting deer varies from estate to estate but it's actually often costly for the landowners to manage the deer. I imagine that any reintroduction is highly unlikely at the moment because the public would need to be onside and the sheep farmers would need to be adequately compensated and would also need to be onside.
AK: How do you think sheep farmers could be persuaded?
TC: Well, I actually think that the big estates that are talking about changing have the right to say to the farmers you will be land managers, which is what they're doing. And I just don't think the sheep farmers will be there given the timescale that it's going to be before there could ever be a considered reintroduction. But I don't think there'll be commercial farming up there at all.
AK: Seeing the wolves basking calmly in the sunshine I was keen to ask John just how much of the stereotypes surrounding wolves is true.
TC: Wolves are omnivores and they are also opportunistic feeders. So if they come across something, I mean, that's one of the reasons they get such a bad name is that if a farmer comes across fallen stock and the wolves are eating it then he blames the wolf. It might be his own domestic dog that's done it or it could just be old age that dropped the thing. But I don't think they'd eat humans. They see us as something to be really, really scared of. They don't want to be anywhere near us really. In America the Indians, the native Americans, have a saying that if you see a wolf it's seen you a thousand times. You only get to see them if they actually want you to see them. Because they can smell human beings about 1.5 to 3 miles away and they can hear us making all the noise we make perhaps up to 10 miles away they don't really need to come anywhere near us. And they actually don't.
AK: So although it's not something to get too excited about just yet the sound or sight of a wolf may become something more familiar to us in the future, but only if you listen very carefully.
GM: That report from Science Media Production MSc student Alison Kirby with the wolves.
Alternative to the reusable tourniquet
Well, now as you can probably tell from the hubbub behind me I've made my way to the senior common room on the South Kensington campus here at Imperial College and cappuccinos are flowing behind me. Just kind of rolling up to lunchtime so lots of clattering of plates as well. I'm here to talk to, and have a coffee with, Christian Fellowes, who's a house officer at the Chelsea and Westminster Hospital. Because Christian you've been working on this really interesting thing: a new kind of tourniquet. First question, why do we need a new tourniquet?
Christian Fellowes: Well, firstly a tourniquet is an instrument that is used to take blood from a patient or put a line in. So, in other words, you just put it round an arm or a leg and therefore make the veins swell up. Well, the reason we feel we need this new product is basically the ones that are used at the moment pose an infection risk. The disposable alternatives, which our product would be competing with, actually are not very practical to use and can be quite expensive as well.
GM: And this is a problem that you've encountered then in your own studies in medicine here at Imperial and you've thought, well, I'm going to do something about this?
CF: Yeah, that's correct. A colleague and I, who's Ryan Kerstein who developed it with me. During our attachments on the wards we'd seen the behaviour of doctors, nurses, and phlebotomists who take blood as well. Even though they were being very judicious in the way they used the old tourniquets they still weren't cleaned properly. They were quite old. We went looking round for alternatives and, again, we found that these weren't ideal as well. So we just took it a stage further really and did a study with someone at Chelsea, Dr Berge Azadian, and confirmed that they are an infection risk and basically took the design from that point.
GM: And we have the fruits of your labour right here on the table. So it comes as a strip of tourniquets. There's a reel of them, basically, and you just pull one off each time you need one. It just basically comes out of this cardboard dispenser here.
CF: The idea is you pull it out. When you take blood you assemble, effectively, a set. This would be part of that and then after you've finished it you can dispose of it in the usual mechanisms in a hospital.
GM: I think we should try it out, obviously minus the taking blood bit if you don't mind because I'm not very good with needles. I can just about handle a tourniquet. So I guess I should just roll my sleeve up here?
CF: That's correct.
GM: We're getting major funny looks from the people in the queue for the coffee bar here. They really are wondering what's going on. Okay, so we're going to put it around my wrist then are we?
CF: Yeah, just around your forearm.
GM: And to describe it as best we can then it is a strip about 20 to 30 centimetres long.
GM: And it's a bit like, you know, people who go to festivals and you get those little wrist band things. It's a bit like one of those but bigger isn't is?
CF: I suppose that was our initial inspiration partly.
GM: So you were wallowing around in the mud at Glastonbury or something and thought, hang on, this could work in medicine?
CF: But the real inspiration was the fact that we were looking for common design techniques and common materials that would obviously reduce the cost of the actual product and therefore make it a viable alternative as a disposable product. And also with the NHS environment where money isn't with us.
GM: So one end loops through the other, I guess? Then there's an adhesive bit that allows you to stick it down?
CF: That's correct. And basically the pressure in the venous system is obviously quite low so you don't need a huge amount of force for it. I mean, here on your arm we can see that the veins themselves are swollen up now prefect for taking blood from them. One problem you always find as well when you're taking blood is that you have one hand you have to release the tourniquet with. And the way this adhesive strip works, and with the excess material on the other end of it, you can just easily pull it to release it and therefore release the pressure and prevent the bleed when you pull the needle out.
GM: And you've released it very quickly and simply and I have to say it's quite a relief to have that pressure off my lower arm. But there you go. It's very, very easy to use. The main point that you're making here is that this is sterile and reduces infection risk then?
CF: As I said before, with the old tourniquets that were reusable and more expensive they go from patient to patient and they basically pick up skin fall bacteria from that and other things in the ward environment on the hands from touching the tourniquets. With cultures of various things including MSRA and things like this on these tourniquets, and obviously by moving it between patients, you potentially harvest this bacteria. Ours being single use. It's not technically a sterile background because it hasn't been sterilised but there'll be a very low bacterial load on it and nothing that's pathogenic, so to speak. So in that way we prevent spreading the bacteria between patients and therefore hopefully reducing infection risk.
GM: And how much do you think this could save the Health Services?
CF: The estimated actual costs of hospital acquired infections is in the region of a billion pounds a year. So actually giving an estimation on that figure is very difficult to do but all measures that are taken obviously to reduce the chance of getting infections is obviously very beneficial from a patient point of view in terms of the morbidity associated with that. And also from an NHS and financial point of view from having to treat those. And obviously we view this actually as not the absolute answer, which it's not, but it's part of an armoury of solutions that are available out there.
GM: And having developed it here you're now marketing it then are you?
CF: The actual marketing is being done under the auspices of Imperial Innovations so they're the ones now going forward with the idea and looking to promote it within the marketplace.
GM: Okay, well, Christian Fellowes, good luck with it and thanks a lot.
And apologies if you were in the senior common room and my bulbous veins put you off your dinner.
Well, that's it for this edition. Join me next month for an exclusive interview with our rector, Sir Richard Sykes. Obviously well known within Imperial but also outside in part thanks to his controversial views on issues like tuition fees. Whatever he tells this podcast I suspect he won't mince his words.
The official podcast of Imperial College is available on the first working day of each month and is a co-production of our Press Office and the Science Communication Group.
Ozgur Buldum is the composer of this theme music. You can hear more of his work at ozgurbuldum.com.
So see you in July. Until then from me, Gareth Mitchell and producer Helena Rant, thanks very much for listening and goodbye.