Podcast: A synthetic eye that 'sees' like a human

Host: Shamini Bundell

Welcome back to the Nature Podcast. This week, making an artificial eye…

Host: Nick Howe

And how disk galaxies formed in the early Universe. I’m Nick Howe.

Host: Shamini Bundell

And I’m Shamini Bundell.

[Jingle]

Host: Shamini Bundell

First up, Darwin once pointed to the human eye as an example of evolution through natural selection that people might find hard to accept. An eye is an incredibly complex organ, but researchers working on artificial vision have managed to craft an eye that mimics our own, complete with a working retina. Reporter Geoff Marsh spoke to the scientists involved to find out more.

Interviewee: Zhiyong Fan

Because I used to see a lot of sci-fi movies, they have the artificial eyes, so I found this technology very interesting. My name is Zhiyong Fan. I’m a professor working in Hong Kong University of Science and Technology.

Interviewer: Geoff Marsh

What’s your favourite sci-fi film with an artificial eye?

Interviewee: Zhiyong Fan

Oh, Star Trek for sure. In 2012, we developed some enabling technologies. Then in 2016, we started with the actual work to fabricate our biomimetic eyes.

Interviewer: Geoff Marsh

We have cameras already with incredible resolution. What’s so special about the layout of the human eye? Why is that something to aim for?

Interviewee: Zhiyong Fan

One of the unique features everybody knows – our eyes are spherical-shaped. So, the spherical shape makes it very easy for the eyeball to move in our eye sockets. Of course, inside our eyeball, we have this hemispherical retina. So, this hemispherical retina also makes our single eye field of view much bigger than our camera, which uses flat image sensors.

Interviewer: Geoff Marsh

And that hemispherical shape, that dome shape, of the retina, am I right in thinking that that is what has posed the greatest fabrication challenge for you and your team?

Interviewee: Zhiyong Fan

Exactly, even now it’s very difficult to fabricate the sensing devices, circuitry and the hemispherical substrate. It’s very difficult. So, that is one of the biggest challenges we encountered in the very beginning.

Interviewer: Geoff Marsh

What are your photo-sensing nanowires? What are they made of?

Interviewee: Zhiyong Fan

So, these are semi-conducting light-sensitive materials. In this work, we used a new material called a perovskite material.

Interviewer: Geoff Marsh

Perovskite?

Interviewee: Zhiyong Fan

Yes. We used perovskite nanowires.

Interviewer: Geoff Marsh

And that’s the same material that’s used in photovoltaics.

Interviewee: Zhiyong Fan

That’s right, the new type of material for photovoltaic technology. So, this material converts the photon into the charges, so these nanowires are light-sensing materials. They are very sensitive to light and their density is around six times higher than the density of the photoreceptors on the human retina, so they have a higher density for sure.

Interviewer: Geoff Marsh

What was the key to managing to embed your photosensor array on that dome shape?

Interviewee: Zhiyong Fan

I guess the key is we achieved the integration and the material fabrication in one step. In the past, people have already tried. People work on nanotechnologies and make small devices. Often, they make the material first, then they transfer the material to somewhere else, so they do things in multiple steps, two steps at least.

Interviewer: Geoff Marsh

So, what your team did differently was you constructed your dome-shaped retina with the photosensing nanowires in that shape. You didn’t start on a flat surface and then try and fold it up.

Interviewee: Zhiyong Fan

That’s right. That’s right. That’s the key.

Interviewer: Geoff Marsh

And so, that allowed you to make that array much denser because you didn’t have to leave gaps between the photosensors to allow for folding.

Interviewee: Zhiyong Fan

Right, right. Another additional feature which makes our technology very unique is our technology is basically an electrochemical photo detector, unlike those solid-state devices. So, it’s a liquid inside. It’s just like how our eyeball has liquid inside. So, every single sensing pixel they’re connected with is liquid. But on the back side, we chose to use liquid metal wires as artificial nerve fibres to connect with the individual sensing elements because they are very flexible and also highly conductive, and the same principle is our human eye.

Interviewer: Geoff Marsh

What could your device ‘see’? How did you test that?

Interviewee: Zhiyong Fan

Well, in principle, we can achieve much higher image resolution than the human eye. So, our human eye, we have a resolution limit. If we look far away, for example, a star far away from us, it might only project a dot on our retina. The size of the dot, if it’s smaller than 3 micrometres on our retina then we can’t even see it, right. But if we use our nanowires and the size of this dot created by the star on our retina, if it’s 1 micrometre then we can see it.

Interviewer: Geoff Marsh

Does that mean then that if in the future someone did have these eyes implanted and they looked up at the night sky that they would see a much richer array of stars?

Interviewee: Zhiyong Fan

Absolutely, definitely you would see more stars. If you don’t look at a star and just look at surroundings, you’ll have a much clearer image and you can see a much further distance.

Interviewer: Geoff Marsh

So, I understand that you’ve put this artificial eye through a series of kind of performance tests.

Interviewee: Zhiyong Fan

So, we tested the sensitivity of our artificial retina. So, we can measure very weak light, and we also measured the response speed of our artificial retina. The response speed is a little bit more than 10 milliseconds, so that is already much faster than the human eye. The human eye will be something like 40 milliseconds, so our artificial retina is four times faster. So, we also measured the imaging functionality, the object projecting onto the retina and then we use a computer to acquire data and reconstruct the object so that we also can demonstrate. So, it has a higher fidelity.

Interviewer: Geoff Marsh

I mean it sounds like if this ever ended up in a human that they would have superpowers.

Interviewee: Zhiyong Fan

That’s right. It would be a super eye, basically.

Interviewer: Geoff Marsh

Yeah, you could take this technology further and change the materials so that they’re sensitive to different parts of the light spectrum, and then wow, that’s a whole new visual world.

Interviewee: Zhiyong Fan

Exactly, all of a sudden, the whole electromagnetic spectrum is open to us.

Interviewer: Geoff Marsh

Ultimately, Zhiyong, what do you suppose this technology will be used for? Is this just going to be a new kind of camera technology or a more realistic robot gadget or do you think that they’re going to find their way into our eye sockets?

Interviewee: Zhiyong Fan

They are two major application scenarios. So, one is help the blind people if they have an illness on the retina. So, we can use our artificial retina to replace that. The other one is humanoid robotics, so their eye also needs to look like human eye. We can use our artificial eye to replace those eyes made by flat technologies.

Interviewer: Geoff Marsh

I suppose another whole challenge which your team have not directly addressed here would be how you could interface such a technology with the human brain. That’s a whole other story.

Interviewee: Zhiyong Fan

Yes, I’m looking for collaborators working on biomedical research to modify our system, for example, to use more biocompatible materials.

Interviewer: Geoff Marsh

Do you think that your enjoying Star Trek has actually led to this discovery?

Interviewee: Zhiyong Fan

Definitely. In my research, I’ve had a lot of crazy ideas from sci-fi movies, so this is just one of them.

Host: Shamini Bundell

That was Zhiyong Fan from Hong Kong University of Science and Technology. If you want to find out more about artificial eyes, then you can find Zhiyong’s paper over in the show notes.

Host: Nick Howe

In a bit, we’ll be hearing about the earliest detected disk galaxy. Right now, though, it’s time for the Research Highlights with Dan Fox.

[Jingle]

Dan Fox

The bioluminescent prey of southern elephant seals have a trick up their sleeves when they’re being hunted – a dazzling flash. By fitting seven seals with light, movement and location sensors, a team of researchers have discovered that some of the bioluminescent creatures, such as squid and lantern fish, that the seals feed on dazzle their attackers with a burst of light. Seals pursuing flashing prey took longer hunting compared to those capturing non-flashing targets, suggesting that bedazzlement is a good defence mechanism. However, one of the seals seems to have wised up to this and devised a trick of its own – subtly twitching its head to cause prey to trigger their flash early and reveal themselves in the darkness. Read the rest of that dazzling discovery over at the Journal of Experimental Biology.

[Jingle]

Dan Fox

Blind people can now see letters if the right part of their brain receives precise electric jolts. People who were sighted before becoming blind typically have damage to their eyes or optic nerve but not to the visual cortex – the region of the brain that processes visual information. In fact, electrically stimulating the visual cortex triggers flashes of light that could allow the brain to create a recognisable picture, but the image is often shapeless. But researchers from the US have harnessed a new technique using an approach similar to tracing a shape on the palm of a hand but instead, transmitting short bursts of electricity in a sequence that mimics the shape of letters. In trials, two individuals who had lost their sight could correctly identify 80% of the letters they were shown, and the researchers think the approach could also be used to trace the outlines of common objects such as houses or cars. Trace out the rest of that research in full at Cell.

[Jingle]

Interviewer: Nick Howe

Next up, there are two main theories about how disk galaxies form. Now, some new observations are providing clues about how this would have happened in the very early Universe. A long time ago, a galaxy formed far, far away. It looked kind of similar to our own Milky Way. It had lots of gas and was a slowly spinning disk. But how do disk galaxies like this form?

Interviewee: Marcel Neeleman

So, there’s two ways that galaxies are forming.

Interviewer: Nick Howe

This is astrophysicist Marcel Neeleman.

Interviewee: Marcel Neeleman

So, one way is that gas accretes onto these systems, so gas from outside falls on to the galaxy and then gets really heated, as hot as possible, and then they slowly fall into the central region and then form stars and then also the external galaxy itself that we can see. Or the second way is that maybe this gas still falls in from outside, but doesn’t get heated as much. It stays cold and then falls directly on to the galaxy. The difference between them is that one will take a lot longer. If gas falls in and gets heated, it takes a long time, so if you can find a galaxy very early on, you might be able to rule that model out.

Interviewer: Nick Howe

These two methods of disk galaxy formation – slowly with hot gas or quicker accumulation of cold gas – are both thought to occur in the Universe. But to work out which method was occurring when, researchers have to look back in time. Simulations have suggested that cold gas method could have been a dominant mode of disk galaxy formation in the early Universe. To confirm this, researchers need to peer back around 12 billion years by looking at something 12 billion lightyears away, and not only detect it but determine it’s a disk galaxy too.

Interviewee: Marcel Neeleman

You need the best telescopes in the world.

Interviewer: Nick Howe

This week in Nature, Marcel has put one of the top telescopes to use. In stunning resolution and detail, he’s observed a disk galaxy that formed only 1.5 billion years after the Big Bang. Universe-wise, that is not a long time, making this the earliest detected disk galaxy so far. The fact that it must have formed so quickly is good evidence for the cold method of galaxy formation.

Interviewee: Alfred Tiley

I was initially very surprised that there was a disk at this epoch.

Interviewer: Nick Howe

This is Alfred Tiley, an astronomer who wasn’t associated with Marcel’s study.

Interviewee: Alfred Tiley

From an observational perspective, everything that we have seen so far, including imaging from the Hubble Space Telescope, studies with our best instrumentation on the ESO Very Large Telescope, all of that was telling us that disks in galaxies start to emerge around 4 billion years after the Big Bang. So, to see evidence for a disk in a galaxy only 1.5 billion years after the Big Bang, that was surprising to me. However, it kind of is in line with theoretical expectations that cold-mode accretion should dominate in the early Universe. I think it’s a very exciting result.

Interviewer: Nick Howe

Alfred was particularly impressed with how well this early disk galaxy was able to be observed.

Interviewee: Alfred Tiley

It’s shown that it’s technically possible to not only detect this light that’s emitted from the gas in galaxies only 1.5 billion years after the Big Bang, but also to detect it with sufficient sensitivity and spatial resolution to really do a proper study of its properties.

Interviewer: Nick Howe

Whilst Alfred does think this newly observed galaxy is good evidence for the cold method of formation being dominant in the early Universe, there are some other possibilities of how it could have formed, such as two non-disk galaxies colliding. So, Alfred would like to see more examples of disk galaxies 1.5 billion years after the Big Bang.

Interviewee: Alfred Tiley

And then you really start to get a clearer picture of, okay, is this systematic across all galaxies at that time or is their galaxy an outlier from the rest of the galaxy population at that time?

Interviewer: Nick Howe

Marcel agrees and his next steps are to try and locate more early disk galaxies.

Interviewee: Marcel Neeleman

We found one so that’s a good start, but we definitely need more to confirm that this is a common way of forming galaxies, but we have good reason to believe that this is more common that we thought it was going to be. So, we found a galaxy that was pretty normal. There was nothing special about this galaxy when you look at it. So, the fact that we found it in this kind of way, that we selected a normal galaxy that doesn’t stick out in any kind of possible way, gives us a good feeling that this was probably a very normal way of forming galaxies.

Interviewer: Nick Howe

Whether or not this cold method of formation was indeed common throughout the early Universe remains to be seen, but both Alfred and Marcel think this is a key first step and shows that observations like this are possible, and that’s something that’s going to be really exciting for the astronomy community who will be keen to peer back into the early Universe.

Interviewee: Alfred Tiley

In a way, given that it’s only one galaxy, that’s even more exciting because I think this is going to push people to try and construct large samples of galaxies at a similar epoch and so I think, in terms of what it’s going to do for astronomy, I think it’s going to really focus the attention of astronomers on galaxy evolution and galaxy formation at a much earlier epoch than has been possible with the previous generation of telescopes and instrumentation.

Interviewer: Nick Howe

That was Alfred Tiley from the University of Western Australia. You also heard from Marcel Neeleman from the Max Planck Institute in Germany. You can check out Marcel’s paper over in the show notes where you’ll also find a link to a News and Views article written by Alfred.

Host: Shamini Bundell

Last up, it’s time for a quick chat about some other non-corona science stories out there. Nick and I have been browsing the Nature Briefing for the past few weeks – that’s Nature’s daily pick of science news and stories – and Nick, what do you want to share this time?

Host: Nick Howe

So, I’ve been looking into a new preventive treatment for HIV. So, at the moment, if you are at high risk of getting HIV, what you would do is take a combination of two drugs every day, but that’s really hard to stick to and it can mean that you leave yourself vulnerable if you forget or you miss a dose or something like that, and so what this is it’s an injection that you take once every two months, and that would be enough to protect you from the virus.

Host: Shamini Bundell

Wow, going from once a day to once every two months would be a massive lifestyle change, and what kind of people is this going to benefit?

Host: Nick Howe

Well, I guess it would benefit those people that are most at risk from HIV, so that might be sex workers, it might be people in relationships with people who are HIV positive who don’t want to run the risk, and there are certain groups of individuals, such as men who have sex with men, that are at higher risk than other groups, so those would be the type of people that it would be for. But as you say, it would be really great to be able to switch from taking two things once a day to something every couple of months. It would be much easier to stick to, and that’s the hope of the researchers.

Host: Shamini Bundell

And so, this is something that they’re still trialling but is looking hopeful.

Host: Nick Howe

Yeah, so this is still in clinical trials and, I must say, it hasn’t been peer reviewed as of yet, and also the trial is being a bit disrupted by the coronavirus outbreak. But it looks good from what they’ve shown so far, and the incidence of HIV in the volunteers is the same as those people taking the current treatment, which are the drugs, and so it looks like it could be a good potential treatment.

Host: Shamini Bundell

Brilliant, well, we’ll keep an eye out for the results of those trials then. My story this week is about bacteria. Woo, love the bacteria, and it’s about bacteria that are apparently hanging out in rocks all over the place, well, under the surface of the oceans, deep within the oceans in the rock, in the Earth’s crust. Somewhere there’s no light, there’s really barely any food, and you really wouldn’t be expecting to find life.

Host: Nick Howe

Those bacteria, they seem to just get everywhere, don’t they? I swear like there is no place that they can’t exist, but I do wonder – as you said, there’s no light, there’s no food – what exactly are these bacteria living on?

Host: Shamini Bundell

Well, it’s not obviously completely unknown that creatures can exist without sunlight and photosynthesis, so of course, deep sea vents are a little fountain of life deep below the surface of the ocean, and some of the bacteria in rocks under the oceans could be sort of getting food that slowly makes its way through. Another possibility is some of them might be getting food from, well, essentially, radioactivity in the rocks, providing that energy that they’re able to convert into a power source.

Host: Nick Howe

Even with radioactivity and things like that, they’re can’t be a lot of energy down there for them to use, so what exactly are they doing?

Host: Shamini Bundell

Well, a lot of them seem to be living and growing, just extremely slowly. So, one of the researchers in this Quantum Magazine article talks about there maybe being a single cell living for 100 or 1,000 years before dividing or reproducing, so potentially a completely different sort of rate of life than us surface folks would ever be able to comprehend. And then studying them might actually be really interesting because if these bacteria can survive with just rock and water, that might give us clues about where to look for life on other planets and what sort of life that could be.

Host: Nick Howe

Well, I guess slow and steady really does win the race then. Thanks for that Shamini. Listeners, if you’d like more short snippets of science just like that we discussed but instead to your email inbox then make sure you check out the Nature Briefing. We’ll put a link to that in the show notes along with links to the articles we discussed.

Host: Shamini Bundell

That’s all for this week. If you want to get into contact with us then you can find us on Twitter – we’re @NaturePodcast – or if you are more email inclined then we’ve got you covered too – we’re podcast@nature.com. I’m Shamini Bundell.

Host: Nick Howe

And I’m Nick Howe. Thanks for listening.