Building medical robots, bacteria-sized |Bradley Nelson |TEDxZurich

We learned of the existence of bacteria over 300 years ago, and we have far more of them in our bodies than human cells, but it was less than 40 years ago when we first realized how they swim. With the discovery of the rotary motor of E. coli in 1973, a motor just 45 nanometers in diameter, some claimed this incredible mechanism as evidence of God, though it is really just a step along the path of evolution. Now we can actually build nanorobots that swim similar to bacteria like E. coli. We're working to use these to deliver drugs to specific locations in the body. E. coli itself is a kind of robot: it has sensors (chemoreceptors), motors, communication along protein guided pathways, and software (DNA). When we look at a bacterium from this perspective, it seems like a machine, even one that hopefully we will be able to duplicate someday. So if bacteria are really just machines, then what are we?

Brad Nelson is the Professor of Robotics and Intelligent Systems at ETH Zürich where his primary research focus is on microrobotics and nanorobotics with an emphasis on applications in biology and medicine. He studied mechanical engineering at the University of Illinois at Urbana-Champaign and the University of Minnesota and robotics at Carnegie Mellon University. He has worked at Honeywell and Motorola and served as a United States Peace Corps Volunteer in Botswana, Africa. He was a professor at the University of Illinois at Chicago and the University of Minnesota before joining ETH in 2002.
Prof. Nelson was named to the 2005 "Scientific American 50," Scientific American magazine's annual list recognizing 50 outstanding acts of leadership in science and technology from the past year, for his efforts in nanotube manufacturing. His lab won the 2007 and 2009 RoboCup Nanogram Competition--both times the event has been held--in which micrometer-size robots competed in soccer. His lab appears in the 2012 Guinness Book of World Records for the "Most Advanced Mini Robot for Medical Use." He serves on the editorial boards of several journals, has chaired several international workshops and conferences, has served as the head of the ETH Department of Mechanical and Process Engineering, as the Chairman of the ETH Electron Microscopy Center (EMEZ), and is a member of the Research Council of the Swiss National Science Foundation.

This talk was given at a TEDx event using the TED conference format but independently organized by a local community. Learn more at https://www.ted.com/tedx

I want to tell you today about three areas
of science and engineering

that I think are converging
in very interesting ways.

I'm a mechanical engineer.

I've been working in robotics
for over 25 years.

I've been in micro/nanotechnologies
for over 15 years.

And over the past decade,
since I've been here in Zurich,

I've been working more closely
with biologists and with medical doctors,

and I think that the technologies
we're working on

and our vision of the future
has some very interesting implications.

But instead of telling you about it,

what I want to show you
is a clip from a Hollywood film

that actually happens
to be almost as old as I am, so ...

(Video) Man: All stations, stand by.

(On stage) (Laughter)

(Video) Man: Right. Inject.

(On stage) "Fantastic Voyage,"
it's a classic.

I love this movie.

Hollywood has two advantages
when they make movies, versus an engineer.

They don't have to worry about physics.

They don't have actually
have to make the things.

What I want to show you now

is an animation actually made for us
by the Discovery Channel.

They visited my lab
about a year and a half ago.

We appeared on one of their shows,

and they put together this concept
of where we're heading.

And what we've been working on
for several years now

have been little, what we call microrobots
that we inject into your eye -

we haven't done it on a human yet,

but we inject it into your eye -

and we use magnetic fields
to guide that device back to the retina

to perform certain retinal therapies,
for instance delivering drugs.

You saw there, over the patient,

the sequence of electromagnetic
coils that we use.

This is in a real pig's eye
that you're seeing right here.

This pig's eye came from the butcher
earlier that morning,

so we didn't harm any animals
ourselves in making this, but -

(Laughter)

What you see is that we're able
to very precisely control that device.

That device is about 0.5 mm in size,

about a millimeter long,
to give you an idea of scale.

And in this next slide,

you'll see on the left is a system
of electromagnetic coils we use.

We do in vivo animal trials with these.

There are eight of these coils,

we call it the OctoMag,

and we control the current
in each one of those very precisely

to guide this device
through the ocular cavity

back to the retina.

You'll see one of our most recent devices
on the fingertip there.

That particular, we call it a microrobot;

it's about 1/3 mm in diameter,
330 microns in diameter.

And our design specs -

the reason we want it to be so thin -

it's about 1.8 mm long -

is that we want it to fit
inside of a 23-gauge needle.

If it fits inside of a 23-gauge needle
and we inject it into your eye,

as we remove that, that puncture wound
doesn't need a suture.

It's relatively non-invasive.

You just put a little topical
anesthetic, and it's done.

All the time to inject drugs to treat
age-related macular degeneration -

that needle, not the microrobots,

I should say.

But that robot that I just showed you,
that you see there on the fingertip,

is the biggest robot we make.

My goal is to make robots that are
about 1000 times smaller than that,

something the size, for instance,
of these E. coli bacteria.

These little rod-shaped bacteria
are about a micron or two long.

That is about 1/100
of the width of a hair.

See those little tails coming off of them?

We'll get to that later, okay?

But before we start talking
about bacteria,

I want to talk a little bit about physics
and what these constraints put on us,

so we're going to do
a simple thought experiment here.

Let's take a cube, okay?

It's a meter on the side.

And I don't need my calculator
to do this calculation.

A meter by a meter by a meter
is a cubic meter, right?

But if I take that cube
and I shrink it to 10 cm -

I shrink it by a factor of 10 -

that calculation changes

because I'm taking a length
by a length by a length,

and all of a sudden, it's become
1/1000th of its original volume,

and so properties that depend on volume -

for instance, mass -

also go down by a factor of 1000.

Now, if I go down another
100 times, to a centimeter,

it's gone down, now, by a million times.

And so volume -

as I said, the weight of it
goes down by a million times,

but also those magnetic forces
we generate on it are also going down

because they scale also
with the mass of the object.

So you might say, "But since
it weighs less, what's the problem?"

But now, let's think
about the surface area of that cube.

It's got six sides,
each side is a square meter.

It's got six square meters on that cube.

Over the volume of one, ratio of six.

But as I go down, that area
is only a length by a length,

and so as I go down each order
of magnitude by a factor of 10,

the importance of surface area
goes up by a factor of 10.

And that causes problems, okay?

I can't make robots

and guide them with magnetic fields
the way I showed you in the eye -

I can't make them any smaller than I have.

So what are some of the implications?

Well, think about a fish
and how a fish swims.

A fish moves its tail back and forth
in a reciprocal motion.

It's pushing the mass of fluid back
and moving itself forward.

It knows Newton's first law, okay?

And so, Geoffrey Taylor,
professor at Cambridge,

thought about this and published
some very important papers in the 1950s,

and he made a little mechanical fish
just to show how it would work in water,

and it swims just the way
you'd think it would.

But if I took that fish

or I took you, and I made you
1,000 or 10,000 times smaller,

and I put you in water,
all of sudden, that water would feel -

even though it has the same viscocity,

the surface effects
or the drag of that water

would be much, much stronger on you.

And so what Geoffrey Taylor did -

this is a video he made in the 1960s -

is he got a vat of something very thick.

I think if you're from the UK,
you know Lyle's Golden Syrup,

and I think that's what
he must have used if you look at it.

So, he took his robot -

it's a little mechanical fish -

put it in there,
and it doesn't go anywhere

because the fluid drag is so strong

and the mass that's pushing back
is so much less than that

that it doesn't move.

And that's the problem
as we go down in scale,

is that we have to rethink
the way things swim

and the way things move.

Well, if you're an engineer
and you don't know how to solve a problem,

what do you do?

You look at nature and think,
"How did nature solve this problem?"

Nature solved this problem
millions, billions of years ago.

We know there's paramecia.

You see the spermatozoa
there on the right?

And they have these special
little hairs on them, these cilia,

these flagella
for the sperm, we call them,

that move in very interesting ways.

Now, nobody knew before 1675
that these things even existed.

Antonie van Leeuwenhoek, in Holland,
was looking in his microscope,

and he was astounded

to see a world of tens of thousands
of little microorganisms swimming,

and he wrote a letter
to the Royal Society the next year.

They verified his results.

People were astounded, what was going on.

And what van Leeuwenhoek
saw in his microscope

was the first time
anybody had ever seen bacteria.

This is a graphic
of one of the rod-shaped ones.

It's about a micron or two long.

And as you look at these
under a microscope -

you saw the one I showed of the E.coli -

you'll notice it has
a little flagella on it.

And as you look at it under a microscope,

what you see is this flagella
seems to be wiggling back and forth,

but if you were able to look at it
from another direction,

you realize it's not wiggling
back and forth; it's actually rotating.

And Howard Berg,

when he was at University of Colorado
in the early 1970s, discovered this,

and what he discovered was astounding:

nature has invented a rotary motor.

Think about it.

Where else in nature
is there a rotary motor?

And Howard has been to our lab
and given us some advice on what to do.

He calls these things
nature's microrobots, okay?

So the body of the bacteria
has sensors on it, chemoreceptors.

Those chemoreceptors communicate
with the motor in the back of it,

to drive it.

That also has software in there.

The software is the chunks
of DNA floating around.

They're just telling it
what parts to make

to keep building the sensors it needs,
the motors it needs, and all that.

And the motor is a fascinating structure.

Since Howard discovered
these bacterial motors in 1973 -

which, by the way some people believe
is evidence of an intelligent designer,

but I don't think
most biologists believe that.

These motors are made
from about 30 to 40 proteins.

They assemble into this structure

that spins up to
160 revolutions per second.

And you see on the right here,
a video from Howard's lab

of fluorescent bacteria
swimming at these speeds.

Remember that the size of these
is a micron or two.

So we looked at this,
and we were thinking,

"What can we learn from this?

How can we take advantage of this?"

So we leveraged some
of our nanotechnology experience

to build something we called
an artificial bacterial flagella.

Now, I can't make that motor yet.

That motor's about
45 nanometers in diameter.

But what I can make is the flagella

of a similar size and shape
that a bacteria has.

And on the front of it there on the left,
you'll see what looks like a head,

and what that is is actually
a little piece of magnet,

and what I can do with that magnet

is I can generate a torque on it
with a magnetic field,

and as I rotate that field -

and these are very, very low fields;

they're about 1000 times
less than an MRI field -

they start to get it to twist,

and as it twists,
it propels itself forward,

just like E. coli do.

To give you an idea of the scale
we're talking about,

here's a scanning electron
micrograph of a human hair;

it's about 100 microns or so in diameter.

There is the size of our smallest ABFs.

They're about 10 microns,
these particular ones.

And this is the size
of a red blood cell, okay?

So we're about double.

Our smallest ones are
about twice the size of a red blood cell.

And here are three of them swimming
together in a sort of swarm behavior.

To me, they look alive.

I get excited when we do this, you know?

(Laughter)

That's why I do robotics.

There's nothing more fun than building
a machine and watching it move.

Now, you'll notice
these will start to go backwards.

I didn't reverse the video;
I just reversed the field.

There's some really interesting
fluid dynamics to be explored here,

and that's pretty interesting.

One exciting thing for us this year
was when we were in the bookstore,

we picked up a copy of
the 2012 Guinness Book of World Records

and discovered that we were
in the Guinness Book of World Records

for the smallest medical robot.

(Audience) Whoo!

Bradley Nelson: Being in the
Guinness Book of World Records is great,

but what I'm really gunning for is,

I want to win a medal
in the next Olympics,

and so we're developing
synchronized swimmers.

(Laughter)

These are interesting -

What's particularly interesting
about these guys

is that they're made out of a polymer.

They're noncytotoxic.

They don't kill cells;

in fact, cells like to grow on them.

And we've developed a new technology

that allows us to make
some fairly arbitrary shapes here.

So in this next little video
I want to show you

is one of our devices.

We put a claw on it,

and so what it can do is go around
and grab these little -

these are 6-micron diameter beads,

so they're about the size
of that red blood cell -

grab those, move them up in 3D,
move them up and down,

and then eventually release them
using these fluidic forces.

We've also been thinking about other,
more serious applications as well.

Here's one of our devices.

We coated it with
a fluorescent molecule called calcein.

This molecule, you're looking at it
in a fluorescent microscope there.

This molecule, actually,

is the same molecular weight
as a lot of chemotherapy drugs.

And on the left, you'll see
some red cells that are stained red.

We discovered as we moved this bacteria
near those cells and touched them with it,

the calcein actually
gets taken up by the cells.

So this allows us, now, to potentially
deliver drugs into individual cells

and target individual cells
with this kind of technology.

The other thing that's cool -

I've only shown you a few,
but we can make armies of these.

We can make them by the thousands.

We can make about one a second.

We make tens of thousands,
put them in suspension.

So I think there's some interesting
possibilities here

for the future of where this can go.

So let's go back to the bacterial motor.

This is a video from
Keiichi Namba's lab at Osaka University.

He and his group have spent years

trying to understand
the exact sequence of proteins,

how they assemble into this rotary motor.

And while I'm not at the point
where I can develop the motor,

I can develop some of these
parts of this device,

and so what we're hoping as we move into
the future and keep going in this area,

we'll learn more and more from nature
at these molecular scales

and be able to build machines
that operate in similar ways

and under similar principles.

I've been very fortunate
to work with some brilliant scientists,

brilliant medical doctors,

and when you're at the ETH,

the Swiss Federal Institute
of Technology here -

you know, I'm an engineer.

I walk the hallways where people
like Conrad Röntgen, who invented X-rays,

Wolfgang Pauli or Albert Einstein were.

It's a humbling experience.

So I take a little bit of comfort

in a quote from a famous
aeronautical engineer from Caltech,

Theodore von Karman,

and von Karman said,

"The scientist describes what is;
the engineer creates what never was."

(Laughter)

Okay. So.

I want to leave you
with one last thought here.

This is from Richard Feynman,
the famous physicist from Caltech,

who said, "What I cannot make,
I do not understand."

(Laughter)

Okay. So thank you very much.

(Applause)