0:00:00.891,0:00:02.854
Every single one of us will lose

0:00:02.854,0:00:06.011
or has already lost something[br]we rely on every single day.

0:00:06.721,0:00:09.222
I am of course talking about our keys.

0:00:09.530,0:00:10.931
(Laughter)

0:00:11.194,0:00:12.210
Just kidding.

0:00:12.210,0:00:15.542
What I actually want to talk about[br]is one of our most important senses:

0:00:15.542,0:00:16.536
vision.

0:00:16.536,0:00:19.260
Every single day we each lose[br]a little bit of our ability

0:00:19.260,0:00:20.594
to refocus our eyes

0:00:20.594,0:00:22.328
until we can't refocus at all.

0:00:22.837,0:00:24.960
We call this condition presbyopia,

0:00:24.960,0:00:27.324
and it affects two billion[br]people worldwide.

0:00:27.554,0:00:29.324
That's right, I said billion.

0:00:29.524,0:00:31.204
If you haven't heard of presbyopia,

0:00:31.204,0:00:34.010
and you're wondering, [br]"Where are these two billion people?"

0:00:34.010,0:00:36.074
here's a hint before[br]I get into the details.

0:00:36.074,0:00:39.217
It's the reason why people wear[br]reading glasses or bifocal lenses.

0:00:39.537,0:00:42.435
I'll get started by describing[br]the loss in refocusing ability

0:00:42.435,0:00:43.774
leading up to presbyopia.

0:00:43.952,0:00:44.968
As a newborn,

0:00:44.968,0:00:48.415
you would have been able to focus[br]as lose as six-and-a-half centimenters

0:00:48.415,0:00:49.636
if you wish to.

0:00:49.636,0:00:52.698
By your mid-20s, you have about[br]half of that focusing power left.

0:00:52.734,0:00:53.950
10 centimeters or so,

0:00:53.950,0:00:56.485
but close enough that you never[br]notice the difference.

0:00:56.485,0:00:57.640
By your late-40s though,

0:00:57.640,0:01:00.008
the closest you can focus[br]is about 25 centimeters,

0:01:00.008,0:01:01.242
maybe even farther.

0:01:01.242,0:01:03.316
Losses in focusing ability[br]beyond this point

0:01:03.316,0:01:05.526
start effecting near-vision[br]tasks like reading,

0:01:05.526,0:01:07.489
and by the time you reach age 60,

0:01:07.489,0:01:09.992
nothing within a meter[br]radius of you is clear.

0:01:10.168,0:01:12.335
Right now some of you[br]are probably thinking,

0:01:12.335,0:01:15.683
that sounds bad but he means[br]you in a figurative sense,

0:01:15.683,0:01:18.622
only for the people that actually[br]end up with presbyopia.

0:01:19.128,0:01:23.578
But no, when I say you, I literally mean[br]that every single one of you

0:01:23.578,0:01:26.429
will someday be presbyopic[br]if you aren't already.

0:01:26.852,0:01:28.237
That sounds a bit troubling.

0:01:28.237,0:01:31.901
I want to remind you that presbyopia[br]has been with us for all of human history

0:01:31.901,0:01:34.724
and we've done a lot of different[br]things to try and fix it.

0:01:34.730,0:01:38.457
So to start, let's imagine[br]that you're sitting at a desk, reading.

0:01:38.699,0:01:40.091
If you were presbyopic,

0:01:40.091,0:01:42.098
it might look a little[br]something like this.

0:01:42.098,0:01:44.840
Anything close by,[br]like the magazine, will be blurry.

0:01:45.196,0:01:46.454
Moving on to solutions.

0:01:46.454,0:01:48.044
First, reading glasses.

0:01:48.285,0:01:50.317
These have lenses[br]with a single focal power

0:01:50.328,0:01:52.408
tuned so that near objects[br]come into focus.

0:01:52.609,0:01:55.322
But far objects necessarily[br]go out of focus,

0:01:55.322,0:01:57.763
meaning you have to constantly[br]switch back and forth

0:01:57.763,0:01:59.535
between wearing and not wearing them.

0:01:59.535,0:02:00.759
To solve this problem

0:02:00.759,0:02:04.026
Benjamin Franklin invented[br]what he called "double spectacles."

0:02:04.241,0:02:06.249
Today we call those bifocals,

0:02:06.249,0:02:09.837
and what they let him do[br]was see far when he looked up

0:02:09.837,0:02:11.459
and see near when he looked down.

0:02:11.459,0:02:14.463
Today we also have progressive lenses[br]which get rid of the line

0:02:14.463,0:02:17.033
by smoothly varying the focal power[br]from top to bottom.

0:02:17.033,0:02:18.445
The downside to both of these

0:02:18.445,0:02:21.244
is that you lose field of vision[br]at any given distance

0:02:21.244,0:02:23.811
because it gets split up[br]from top to bottom like this.

0:02:23.811,0:02:25.180
To see why that's a problem,

0:02:25.180,0:02:27.975
imagine that you're climbing[br]down a ladder or stairs.

0:02:28.166,0:02:31.325
You look down to get[br]your footing but it's blurry.

0:02:31.691,0:02:33.083
Why would it be blurry?

0:02:33.297,0:02:36.567
Well, you look down[br]and that's the near part of the lens,

0:02:36.567,0:02:39.275
but the next step was past arm's reach,

0:02:39.275,0:02:41.104
which for your eyes, counts as far.

0:02:41.274,0:02:44.143
The next solution I want to point out[br]is a little less common

0:02:44.143,0:02:46.757
but comes up a lot in contact lenses[br]or LASIK surgeries,

0:02:46.757,0:02:48.070
and it's called monovision.

0:02:48.070,0:02:50.590
It works by setting up[br]the dominant eye to focus far

0:02:50.590,0:02:52.114
and the other eye to focus near.

0:02:52.114,0:02:54.830
Your brain does the work[br]of intelligently putting together

0:02:54.830,0:02:56.816
the sharpest parts from each eye's view,

0:02:56.816,0:02:59.004
but the two eyes see[br]slightly different things,

0:02:59.004,0:03:01.620
and that makes it harder[br]to judge distances binocularly.

0:03:01.620,0:03:03.150
So where does that all leave us?

0:03:03.150,0:03:06.001
It seems like we've come up[br]with a lot of different solutions

0:03:06.001,0:03:08.338
but none of them quite restore[br]natural refocusing.

0:03:08.338,0:03:10.334
None of them let you[br]just look at something

0:03:10.334,0:03:11.738
and expect it to be in focus.

0:03:11.738,0:03:12.736
But why?

0:03:12.840,0:03:13.984
Well, to explain that

0:03:13.984,0:03:16.693
we'll want to take a look[br]at the anatomy of the human eye.

0:03:16.693,0:03:19.877
The part of the eye that allows us[br]to refocus to different distances

0:03:19.877,0:03:21.398
is called the crystalline lens.

0:03:21.654,0:03:25.644
There are muscles surrounding the lens[br]that can deform it into different shapes,

0:03:25.644,0:03:27.794
which in turn changes its focusing power.

0:03:27.794,0:03:29.913
What happens when someone[br]becomes presbyopic?

0:03:30.213,0:03:32.399
It turns out that[br]the crystalline lens stiffens

0:03:32.399,0:03:35.078
to the point that it doesn't[br]really change shape anymore.

0:03:35.078,0:03:39.080
Now, thinking back[br]on all the solutions I listed earlier,

0:03:39.080,0:03:42.686
we can see that they all have[br]something in common with the others

0:03:42.686,0:03:44.311
but not with our eyes,

0:03:44.311,0:03:46.204
and that is that they're all static.

0:03:46.290,0:03:49.094
It's like the optical equivalent[br]of a pirate with a peg leg.

0:03:49.094,0:03:51.990
What is the optical equivalent[br]of a modern prosthetic leg?

0:03:52.291,0:03:55.524
The last several decades have seen[br]the creation and rapid development

0:03:55.524,0:03:57.927
of what are called "focus-tunable lenses."

0:03:58.333,0:04:00.020
There are several different types.

0:04:00.020,0:04:01.813
Mechanically-shifted Alvarez lenses,

0:04:01.813,0:04:03.447
deformable liquid lenses

0:04:03.447,0:04:05.875
and electronically-switched, [br]liquid crystal lenses.

0:04:05.959,0:04:07.686
Now these have their own trade-offs,

0:04:07.686,0:04:10.287
but what they don't skimp on[br]is the visual experience.

0:04:10.287,0:04:13.618
Full, field-of-view vision that can be[br]sharp at any desirable distance.

0:04:13.618,0:04:14.619
OK, great.

0:04:14.619,0:04:16.193
The lenses we need already exist.

0:04:16.193,0:04:17.293
Problem solved, right?

0:04:17.786,0:04:19.002
Not so fast.

0:04:19.156,0:04:22.054
Focus-tunable lenses add a bit[br]of complexity to the equation.

0:04:22.162,0:04:25.981
The lenses don't have any way of knowing[br]what distance they should be focused to.

0:04:25.981,0:04:28.887
What we really need are glasses[br]that, when you're looking far,

0:04:28.887,0:04:30.001
far objects are sharp,

0:04:30.001,0:04:31.153
and when you look near,

0:04:31.153,0:04:33.924
near objects come into focus[br]anywhere in your field of view

0:04:33.924,0:04:35.908
and without you having to think about it.

0:04:35.908,0:04:38.375
What I've worked on[br]these last few years at Stanford

0:04:38.375,0:04:40.914
is building that exact intelligence[br]around the lenses.

0:04:40.914,0:04:44.480
Our prototype borrows technology[br]from virtual and augmented reality systems

0:04:44.480,0:04:45.917
to estimate focusing distance.

0:04:45.917,0:04:49.487
We have an eye tracker that can tell[br]what direction our eyes are focused in.

0:04:49.487,0:04:50.488
Using two of these,

0:04:50.488,0:04:52.691
we can directly triangulate[br]your gaze direction

0:04:52.691,0:04:53.864
to get a focus estimate.

0:04:53.864,0:04:54.864
Just in case though,

0:04:54.864,0:04:57.483
to increase reliability[br]we also added a distance sensor.

0:04:57.483,0:04:59.820
The sensor is a camera[br]that looks out at the world

0:04:59.820,0:05:01.420
and reports distances to objects.

0:05:01.420,0:05:04.420
We can again use your gaze direction[br]to get a distance estimate

0:05:04.420,0:05:05.599
for a second time.

0:05:05.599,0:05:07.580
We then fuse those two distance estimates

0:05:07.580,0:05:10.075
and update the focus-tunable[br]lens power accordingly.

0:05:10.266,0:05:13.592
The next step for us was of course[br]to test our device on actual people.

0:05:13.592,0:05:16.680
So we recruited about 100 presbyopes[br]and had them test our device

0:05:16.680,0:05:18.392
while we measured their performance.

0:05:18.392,0:05:21.719
What we saw convinced us right then[br]that autofocals were the future.

0:05:21.719,0:05:23.634
Our participants could see more clearly,

0:05:23.634,0:05:25.023
they could focus more quickly

0:05:25.023,0:05:28.031
and they thought it was an easier[br]and better focusing experience

0:05:28.031,0:05:29.461
than their current correction.

0:05:29.461,0:05:30.457
To put it simply,

0:05:30.457,0:05:31.603
when it comes to vision,

0:05:31.603,0:05:34.811
autofocals don't compromise[br]like static corrections in use today do.

0:05:34.811,0:05:36.758
But I don't want to get ahead of myself.

0:05:36.758,0:05:39.500
There's a lot of work[br]for my colleagues and me left to do.

0:05:39.500,0:05:41.948
For example, our glasses are a bit --

0:05:41.948,0:05:42.947
(Laughter)

0:05:42.947,0:05:44.462
bulky, maybe?

0:05:44.462,0:05:47.698
And one reason for this[br]is that we used bulkier components

0:05:47.698,0:05:50.492
that are often intended[br]for research use or industrial use.

0:05:50.492,0:05:52.754
Another is that we need[br]to strap everything down

0:05:52.754,0:05:56.485
because current eye-tracking algorithms[br]don't have the robustness that we need.

0:05:56.776,0:05:58.051
So moving forward,

0:05:58.051,0:06:00.556
as we move from a research[br]setting into a start-up,

0:06:00.556,0:06:02.618
we plan to make future autofocals

0:06:02.618,0:06:05.171
eventually look a little bit more[br]like normal glasses.

0:06:05.171,0:06:06.168
For this to happen,

0:06:06.168,0:06:08.554
we'll need to significantly improve

0:06:08.554,0:06:10.667
the robustness[br]of our eye-tracking solution.

0:06:10.667,0:06:14.539
We'll also need to incorporate smaller[br]and more efficient electronics and lenses.

0:06:14.919,0:06:17.211
That said, even with[br]our current prototype,

0:06:17.211,0:06:19.762
we've shown that today's[br]focus-tunable lens technology

0:06:19.762,0:06:23.345
is capable of out-performing[br]traditional forms of static correction.

0:06:23.345,0:06:24.894
So it's only a matter of time.

0:06:25.134,0:06:27.158
It's pretty clear that in the near future,

0:06:27.164,0:06:30.168
instead of worrying about which pair[br]of glasses to use and when,

0:06:30.168,0:06:32.725
we'll be able to just focus[br]on the important things.

0:06:33.687,0:06:34.686
Thank you.

0:06:34.686,0:06:36.438
(Applause)