- I don't remember.

- [Matt] This is the Google Pixel four.

It has a 5.7 inch OLED display, a Snapdragon 855 processor,

and it can track your pupils on a sub millimeter level

with a reported median error of three and a half percent.

Okay, this video isn't really about a three year old phone.

If you wanna watch that review, we did one.

You can still check it out if you want to.

This video is about the camera in this phone, however.

And how researchers at UCSD are using it to do more

than just unlock it with their face.

Instead, they're seeing if it can measure your pupil size,

which they hope could potentially screen

for neurological diseases like Alzheimer's.

This is all really interesting to me

because when we talk about health and tech,

usually the conversation is around newer devices

like wearables, which not everyone has.

So, I wanted to find out more about this technology,

which again, is on an older phone,

and exactly what other health monitoring can be done

with the device you may already have.

- My name is Colin Barry.

I'm a second year PhD student at UC San Diego

in the electrical and computer engineering department.

And most of our work here is focused

around using ubiquitous technology.

So smartphones, wearables,

and things that we already have around us

to do health based computation.

- So I'm just wondering,

how does pointing a smartphone camera in your eye

tell you anything about neurological diseases,

like Alzheimer's?

- [Colin] It can be difficult to understand

why the eye would be, you know,

a metric for different neurological diseases.

And, first I guess to tell you a little bit

about the measurement.

- That measurement is really just the size of your pupil.

Now, if you don't know,

the pupil is the dark center of your eye

that expands and contracts.

It's just like a camera iris

that lets in more or less light,

depending on how wide the iris is open.

But our pupils can tell more of a story

than just how bright it is outside.

- [Colin] So it's not just a quick snapshot of the eye.

Actually, it's recording your response to specific tasks.

These are correlated with neurological diseases

because of the connections within the brain

between certain brain regions associated

mostly with fight or flight response.

And so what the pupil is doing

is basically giving scientists sort of a window

into the brain and seeing how we can not basically

identify different traits or different cognitive impairments

by the dysfunction or the function of your pupil response.

- In other words,

how your pupil responds to certain tasks

can tell scientists a lot about your cognitive health.

So how does this app allow them to do that?

- The way the app works is that it's going to be using

the facial recognition system,

because it includes near infrared cameras.

So these cameras are recording in a non-visible range.

It's not the same colors and things

that we and you will see when we look around.

And this is especially important for pupil measurements,

because some people have, you know, dark brown eyes

because of the melanin in their eyes.

And the difference between

their black people and their dark iris

is gonna be very minimal in the visible spectrum.

But when you look at it in the infrared spectrum,

it's, you know, almost basically,

black and white between the two

and that's how we're able to get

these sub millimeter measurements of the people.

- There's actually a lot of work going on at UCSD

when it comes to unlocking the technology in your smartphone

for health applications.

- There's a huge sort of library of different measurements

that you can take with a smartphone

and a lot of work needs to be done

to have these measurements sort of validated

and have them not only validated by themselves,

but validated across many smartphone models.

- Many smartphone models is a key point here too,

because not all smartphones have IR cameras.

The Pixel actually lost them after the four.

Now Vera and I aren't in the age demographic

that Collins research group is studying,

but he was still nice enough to lead us through

our own pupil exams, which was pretty easy,

for the most part.

- [Woman] Six, nine, one.

If you need to, blink now. Repeat.

- Six, nine, one, seven, four, two, eight.

- Maybe you should do it in creation.

- This is getting harder.

So the app isn't designed to show us what it's recording.

Colin collects the data and reads it back at the lab,

which will look like this.

A chart showing our pupil changing signs over time,

and this somewhat creepy video of my eyeball.

Now, if that doesn't seem like much,

getting this information is actually pretty hard.

- So generally the pupillometry measurements

are done in clinical lab in very controlled settings

with specialty devices called pupillometers.

They can cost upwards of nine or $10,000,

which is a huge limiting factor,

especially for a lot of clinics.

- Now, Colin isn't suggesting

that your Pixel four is going to replace a $10,000 machine.

There's still a long way to go to find out

if this approach is accurate enough to be useful.

What he is doing here is called a feasibility study.

It's meant to figure out what could be possible.

- The vision is that hopefully people can use these,

you know, smartphone based measurements

to have people take their measurements at home,

outside of the clinic

and then the doctors can use that to understand

whether they need to take more measurements in the clinic.

- That accessibility is important.

Not everyone has access to a clinic with this machine.

But a lot of people do have smartphones, which is good.

But still not a perfect solution.

(somber music)

- Yeah, so smartphones are gonna be more

widely used than wearables.

So, you get some potential for more access there.

- [Matt] Nicole Wetsman is a health reporter

here at The Verge who's covered tons of stories

on the intersection of healthcare and technology.

- I think that digital divide issues

are always gonna be relevant when you're looking at

adding health products to a digital device,

just because someone has a smartphone,

doesn't mean that they have the background and experience

to be able to use it to the fullest extent

of its capabilities.

When you see this happening with

like the telemedicine space, during the pandemic,

just because someone has a computer in their home

doesn't mean that they're gonna be able to access

a telehealth service maybe that's because

they don't know how to download the application

or the application is not available

in the language that they speak,

or they don't have reliable internet access,

or they don't have a quiet space

to make that kind of happen.

- When we're meeting with participants,

a lot of what we do is actually just explaining

how to use the phone and you know,

how to touch on a touchscreen, how to turn the phone on.

- This is something that Colin and Nicole

are acutely aware of.

There are a lot of health apps out there

claiming to do a lot of things,

but there's no guarantee that the public is going to know

how to use them correctly.

And some populations,

they'll get left out of using them completely.

That's why the type of research

that Colin and other scientists do is important.

This isn't about rushing to get the latest feature out

on the newest flagship device.

It's more about

how do we take these pretty incredible devices

that we already have and find out how much good

we can do with them for the most people.

Now we'll stop to wait and see if Colin

and this team's research pans out

and even if it does, that's not the end of the story.

Like, does a company like Google,

decide to buy this research or recreate it?

Or does this actually become developed

as a legit clinical tool used by doctors

in conjunction with traditional devices?

They're both pretty intriguing options

that could grant more access to more people.

- Yeah, I think that simplification of screening

can be useful, you know,

leveraging the technology at our disposal,

you know, the smartphones are really ubiquitous

and if those can have important health applications,

then it's, you know, I think really useful to try

to investigate that to the fullest extent that we can,

like, let's see what we can do with this.

- Huge thanks to Colin Barry from UCSD

for sharing his research with us.

They're doing really cool things over there

with the tech that we use every day.

And thanks to Nicole for giving us the greater context

on healthcare and tech.

Be sure to check out her writing on theverge.com.

So, yeah.

Pixel four, still impressing, three years later.

Not bad.