Something new is on the horizon in the field of rehabilitation.

This patient suffered a stroke, and as a result, can barely move her right hand.

To aid in her rehabilitation, the facility has set up a robot hand.

Move.

Release.

In the instant when she visualizes moving her hand in her mind,

the robot uses bio-electric signals to move the paralyzed hand as envisioned.

In general, the first three months after a stroke are the most effective period for recovery of function.

But the hand is a challenge.

Only 30 to 40% of patients recover enough function to eat normally.

On her fourth day of rehabilitation using the device, a change occurred.

My index finger suddenly moved upward.

It was such a surprise. I didn't expect that.

I said, "look everybody, it moved!"

It's like the part of my brain
that was gone came back to life.

It feels like my brain and
my finger are reconnected.

Like the connection was restored.
I guess that's how it feels.

First available in 2022, this rehabilitation system uses cyborg technology that's recently gotten global attention.

It was developed by a startup entrepreneur and innovator in robotics, Masahiro Kasuya.

He's developed a number of robots using cyborg technology,

like reading biosignals from human muscles to mimic their function.

People are at the heart of it;
their desires, things they want to do.

Using cyborg technology to enable this
is what we would like to do.

This time, we look at how the fusion of human and machine, will shape our future.

A hospital specializing in neurosurgery located in Muroran, Hokkaido.

A region with an aging population, the facility here treats over 800 stroke patients a year.

Strokes, in which the brain is damaged by blocked or ruptured blood vessels, often result in partial paralysis.

So, rehabilitation to recover function is vital.

Okay, make a fist once again.

This patient suffered a stroke more the 20 days ago and has partial paralysis on the left side.

Her left hand is almost entirely immobile.

Ordinarily when you move your hand, nerve signals from the brain

travel down the spinal cord and through the peripheral nerves to the muscles.

A contracting muscle produces a myoelectric signal, a current, in response to this stimulus.

When nerve cells in the brain are damaged by a stroke,

the signals they generate become weak, and aren't properly transmitted.

Paralysis is the result.

This finger training rehabilitation system detects faint biosignals and interprets them.

To accomplish this, three electrodes are positioned on the forearm.

Okay, squeeze. That's it, good.

A completely immobile hand can now move slightly.

The robot assists movement by reading the electrical signals from the brain.

When attempting to move the hand,

a weak signal is sent to the arm through nerve pathways, causing a myoelectric discharge.

This is detected by the electrodes and analyzed.

When the desired movement is determined...

The robot hand moves within a few tenths of a second.

Seeing and feeling this movement, is believed to be a form of biofeedback that promotes re-learning in the brain.

With repeated stimulation, the brain adapts.

Nerve cells surrounding the damaged area acquire new function, creating a sort of "detour," or new neural pathway.

It's thought that, as a result, patients will be able to re-learn to move their hands.

With a therapist's help, patients do exercises like coordinated breathing while moving the paralyzed hand.

Here we go... Okay, okay, okay.

However, since they can't observe patient biosignals themselves, getting the timing right is difficult.

With the device, patients can observe the movement of the hand on the screen.

The detected myoelectrical waveform and its type are displayed.

Open.

In the case of a healthy individual, moving the hand according to the device...

The myoelectric signal for opening the hand is green,

when relaxed it's colorless, and for gripping it's blue.

The signal appears as a continuous line during myoelectric activity.

Open.

Relax.

In the case of a patient with paralysis, even though they are unable to move,

myoelectric signals are still present.

Although the waveforms are small and may be abnormal.

The device interprets the patient's command and strengthens the signal.

Open.

Relax.

The robot hand then moves according to the patient's command.

It gives them the sensation of controlling the hand via the onscreen display and corresponding movement.

This patient is now capable of slight movement.

Open. Okay, open.

It took some time, but the signal was generated.

It's possible for the system to display subtle changes in these biosignals.

It's thought that repeating this type of rehabilitation

in which image and movement are linked helps the brain to re-learn how to move the body.

After rehabilitation, the therapist checks the condition of the hand.

Yes! That's it!

It is moving, just a little.

The hand indeed shows signs of recovery.

Improvement seemed hopeless,
so I guess this is progress.

Many who've had strokes can't
remember how it felt to move,

or they have trouble imagining it.

Being able to visualize if biosignals
are actually being read or not,

having it on the screen is a big plus.

But how is the device able to read these biosignals in such detail?

The answer is AI technology; a field which has been evolving rapidly in recent years.

System developer, Masahiro Kasuya.

With PhDs in robotics and AI engineering,

he's been involved in every step of the process, from design to software development.

The rehabilitation system is the first product from his company to see practical use.

Real-world usability is key.

It won't be used by AI specialists,
but by doctors and therapists.

Making it easy for them to use
was something we focused on.

With conventional biosignal processing it had been difficult to identify the type of signal,

even though it was possible to determine the strength by the size of the waveform.

Kasuya examined myoelectric data from thousands of people.

Depending on the action, such as making a fist or bending the wrist, frequency and amplitude changes.

This type of data was then processed by machine-learning.

A unique algorithm was developed that, by capturing the myoelectric currents in the arm,

was able to identify the desired movement of the hand based on subtle differences in the waveform.

Open.

What movement is the human trying to make?

The AI system they developed boasts high-accuracy real-time analysis,

making it possible to easily distinguish the desired myoelectric signal from unwanted noise.

In general, electrodes must be placed accurately to detect the myoelectric signals...

Basically, anywhere is okay.

Ordinarily it's quite difficult,
but with the rough outline defined

the AI can fine tune it later.

Using biosignals, you can move it
in the way you move your own body.

So, instead of being an object,
it becomes like a part of you.

This concept is what interests me.

Around 2 months ago, Kinue Kayama suffered a stroke, becoming unable to grasp or pinch with her right hand.

26 days after onset, she began rehabilitation with the device for one hour every day.

And on the fourth day, she was able to move her fingers.

One month after using the device.

Kayama is now transitioning to rehabilitation, picking up and releasing objects, unassisted by the robot.

She'll soon be leaving the hospital.

It's the first time I've been able
to do such delicate work.

It's important.

And when I go home, I'll keep at it
to get back even more function.

I'll do my best for another 10 years
at least, and try to make the most of it.

Recently, this type of so-called "neurorehabilitation" has seen real-world implementation and has attracted much interest.

The feeling that they're doing it
themselves is very important.

I guess it's biofeedback.

Do this and it moves, do that it opens.
They develop an internal sense for it.

This kind of neurorehabilitation,
as we call it, is extremely important.

The treatment may also be effective for patients whose muscles have stiffened over time after the onset of a stroke.

This patient suffered a stroke 6 months earlier.

She began rehabilitation after receiving a botulinum treatment to loosen the muscles.

Doing so, movement returned to the immobile fingers, allowing her to wring out this dish cloth.

It was previously believed that loss of function caused by damage to the brain from a stroke was irreversible.

But recent studies have shown that the brain can adapt, making it possible to restore lost function.

In addition to biosignal interpretation, the system developed by Kasuya has another technological advantage:

a robotic control mechanism capable of sophisticated movements.

How best to move a robotic arm?

Kasuya focused on the human body.

Closely studying the physiology of muscles and tendons.

The hand moves here, but the muscles that
move it are actually here.

We began with the physiology
of how the hand actually moves.

And this is the first version of
the prosthetic hand we developed.

His design imitates the human hand, reproducing its movements using wires in place of muscles and tendons.

Being wire driven, means no large
motors are needed in the hand.

This would make it too heavy.

Wire drive lets you move while gripping,
and is significantly lighter.

But existing technology made it difficult to accurately reproduce the hand's delicate movements using wires.

So Kasuya evaluated the material properties of the wire, devising pathways for it to move in.

The control software was also built from scratch.

Repeated testing was performed with the hand worn by a disabled person.

With the same size and weight as the human hand,

it is now capable of performing both powerful and delicate movements, unimaginable with a conventional prosthetic hand.

Using the prototype hand, they entered a competition.

The Cybathlon is an international competition in which people with disabilities compete wearing state-of-the-art prosthetic arms.

Their experience competing against the best in the world at the competition

made Kasuya feel that his company's technology could find global acceptance.

Kasuya is also working on research and development in areas outside of medicine.

This is his "avatar robot."

When it was unveiled in 2018, it attracted worldwide attention.

The avatar robot remotely reproduces the movements of a human operator with high precision.

Even bending and twisting all five fingers, Kasuya's actions are reproduced.

Full freedom of movement.

This robot too is wire driven.

The smooth human-like movements of the hands in particular stunned researchers around the world.

With five fingers, it can use tools designed for humans to perform a variety of tasks.

And even when in a remote location from the operator, movements are reproduced with minimal latency.

Use in the difficult-to-reach environment of space is one anticipated application.

For example, on a space station or the surface of the moon,

work once done by astronauts could be done from the ground.

We could be here on earth while working in space.

To make this happen, Kasuya worked with JAXA, the Japanese space agency,

to simulate work done by astronauts on the space station.

These developments also led to the further evolution of a hand-only avatar robot.

The accuracy of fine movements has also been improved.

Allowing the robot to scoop up these fine grains with a spoon.

It can even lay individual portions out for wrapping.

It can pinch these soft cotton balls without crushing them.

And subtle power adjustments are handled easily.

But it's also capable of heavy lifting, like this 5-kilogram weight.

The operator's motions are reproduced with incredible accuracy.

Avatar robots capable of complex tasks may someday be found at construction sites and other places unsafe for humans.

Being controllable from anywhere on earth, they could fundamentally change the way we work.

Such commercial applications are very close to being a reality.

Doctors have used it for patients,
and gotten feedback.

Yes. And they specifically mentioned
the system in their questions.

Is that so? The questioners?

That's real brand recognition. Fantastic!

Cyborg technology could bring revolutionary changes to our daily lives.

But where did Kasuya find the inspiration for his endeavors?

Loving both robots and making things from an early age, he won prizes for his inventions.

This was the title of his elementary school graduation essay at age 12:

"My Dream Is to Be a Medical Inventor."

I wanted to do medicine and engineering.
I didn't know the word, "cyborg," yet.

A doctor who's also an inventor
is really what I imagined I'd be.

I had a feeling for what I wanted to do
but I had trouble putting it into words.

He decided to pursue cyborg engineering at age 14.

Transfer the data.

He fell in love with an anime depicting the fusion of humans and machines.

He even drew a sketch of a drive system for a powered suit to strengthen human capabilities.

This part labeled AM is
the artificial muscle unit.

And the basic circuit that drives it.

So where the power comes from and
how it's controlled are also included.

I used to think about things like this
all the time back then.

As an undergraduate he attended the Waseda University school of science and engineering.

Working part-time to save money, he incredibly was able to make the power suit he'd imagined a reality;

a device that assists the elbow, making it easier to lift heavy objects.

In 2013 as a graduate student at The University of Electro-Communications,

he started a business with fellow research students.

Maybe reroute these wire pathways...

His dream of a practical application for cyborg technology became a reality.

Having stunned the world with his rehabilitation system and robot avatar,

he's now moving on to the next phase of his efforts.

In 2020, Kasuya established a new base for development in Minamisoma, Fukushima.

Here, he and his team are developing state-of-the-art avatar robots.

The base was set up as an environment in which repeatable testing and experimentation could be conducted,

allowing the development of avatar robots to proceed at a rapid pace.

A one-of-a-kind facility for robot performance evaluation and operator training is also located nearby.

A national project to establish the foundation for a new industry,

the vast grounds include bridges and tunnels, flooded townscapes and a variety of other testing environments.

They're currently testing a caterpillar track system to allow an avatar robot to travel over rough terrain.

Here, they check the function of the prototypes and conduct demonstration trials.

The local area was hard hit by the 2011 Great East Japan Earthquake.

This famously caused the accident at the Fukushima Daiichi Nuclear Plant.

Its decommissioning has become a long and difficult task.

Kasuya is also participating in the development of a robot to aid in this ongoing process.

The plan is to use remotely controlled robots for analysis of radioactive materials.

Our robot hand doesn't just do one thing.

Like a human hand, it can do multiple tasks.
This means it can work quite efficiently.

For example, it could allow workers
to work without exposure to radiation.

We're very close to achieving this.

We're at the stage of development
where such things will be possible.

Avatar robots working in place of humans.

Through cyborg technology, the fusion of humans and machines, a new world of possibilities is on the horizon.

Our ultimate vision is that anyone,
regardless of mobility, or physical condition

can live in the way that they want to.

In addition to prosthetic hands,
we want to develop other parts,

essentially for the whole body.

Those are our current plans.

This massive diagram lays out various perspectives on cyborg technology,

including areas such as religion and culture.

Life is changing rapidly, so how can we
make that experience consistent

with what we've known up to now?

I don't think cyborg technology
should be forced on people,

but an option; some use it, some don't.

How humans deal with this technology,
and how we should make use of it,

this needs to be clearly defined
as we move forward.

What will a future shaped by cyborg technology look like?

The real-world efforts of Kasuya and others will likely provide the answer.