sponsored links
TEDGlobal 2012

Max Little: A test for Parkinson's with a phone call

June 26, 2012

Parkinson’s disease affects 6.3 million people worldwide, causing weakness and tremors, but there's no objective way to detect it early on. Yet. Applied mathematician and TED Fellow Max Little is testing a simple, cheap tool that in trials is able to detect Parkinson's with 99 percent accuracy -- in a 30-second phone call.

Max Little - Applied mathematician
Max Little is a mathematician whose research includes a breakthrough technique to monitor – and potentially screen for – Parkinson's disease through simple voice recordings. Full bio

sponsored links
Double-click the English subtitles below to play the video.
So, well, I do applied math,
00:15
and this is a peculiar problem
00:18
for anyone who does applied math, is that
00:19
we are like management consultants.
00:21
No one knows what the hell we do.
00:23
So I am going to give you some -- attempt today
00:25
to try and explain to you what I do.
00:27
So, dancing is one of the most human of activities.
00:30
We delight at ballet virtuosos and tap dancers
00:33
you will see later on.
00:37
Now, ballet requires an extraordinary level of expertise
00:38
and a high level of skill,
00:41
and probably a level of initial suitability
00:43
that may well have a genetic component to it.
00:46
Now, sadly, neurological disorders such as Parkinson's disease
00:48
gradually destroy this extraordinary ability,
00:51
as it is doing to my friend Jan Stripling, who was
00:53
a virtuoso ballet dancer in his time.
00:56
So great progress and treatment has been made over the years.
00:59
However, there are 6.3 million people worldwide
01:02
who have the disease, and they have to live with
01:05
incurable weakness, tremor, rigidity
01:08
and the other symptoms that go along with the disease,
01:11
so what we need are objective tools
01:13
to detect the disease before it's too late.
01:15
We need to be able to measure progression objectively,
01:18
and ultimately, the only way we're going to know
01:21
when we actually have a cure is when we have
01:24
an objective measure that can answer that for sure.
01:26
But frustratingly, with Parkinson's disease
01:29
and other movement disorders, there are no biomarkers,
01:32
so there's no simple blood test that you can do,
01:34
and the best that we have is like
01:37
this 20-minute neurologist test.
01:39
You have to go to the clinic to do it. It's very, very costly,
01:41
and that means that, outside the clinical trials,
01:43
it's just never done. It's never done.
01:46
But what if patients could do this test at home?
01:49
Now, that would actually save on a difficult trip to the clinic,
01:52
and what if patients could do that test themselves, right?
01:54
No expensive staff time required.
01:58
Takes about $300, by the way,
02:00
in the neurologist's clinic to do it.
02:01
So what I want to propose to you as an unconventional way
02:03
in which we can try to achieve this,
02:06
because, you see, in one sense, at least,
02:08
we are all virtuosos like my friend Jan Stripling.
02:09
So here we have a video of the vibrating vocal folds.
02:13
Now, this is healthy and this is somebody making speech sounds,
02:16
and we can think of ourselves as vocal ballet dancers,
02:20
because we have to coordinate all of these vocal organs
02:23
when we make sounds, and we all actually
02:25
have the genes for it. FoxP2, for example.
02:28
And like ballet, it takes an extraordinary level of training.
02:30
I mean, just think how long it takes a child to learn to speak.
02:33
From the sound, we can actually track
02:35
the vocal fold position as it vibrates,
02:38
and just as the limbs are affected in Parkinson's,
02:40
so too are the vocal organs.
02:42
So on the bottom trace, you can see an example of
02:45
irregular vocal fold tremor.
02:47
We see all the same symptoms.
02:49
We see vocal tremor, weakness and rigidity.
02:50
The speech actually becomes quieter and more breathy
02:53
after a while, and that's one of the example symptoms of it.
02:55
So these vocal effects can actually be quite subtle,
02:57
in some cases, but with any digital microphone,
03:00
and using precision voice analysis software
03:03
in combination with the latest in machine learning,
03:06
which is very advanced by now,
03:08
we can now quantify exactly where somebody lies
03:10
on a continuum between health and disease
03:13
using voice signals alone.
03:16
So these voice-based tests, how do they stack up against
03:18
expert clinical tests? We'll, they're both non-invasive.
03:21
The neurologist's test is non-invasive. They both use existing infrastructure.
03:23
You don't have to design a whole new set of hospitals to do it.
03:27
And they're both accurate. Okay, but in addition,
03:30
voice-based tests are non-expert.
03:32
That means they can be self-administered.
03:35
They're high-speed, take about 30 seconds at most.
03:37
They're ultra-low cost, and we all know what happens.
03:40
When something becomes ultra-low cost,
03:42
it becomes massively scalable.
03:45
So here are some amazing goals that I think we can deal with now.
03:47
We can reduce logistical difficulties with patients.
03:51
No need to go to the clinic for a routine checkup.
03:53
We can do high-frequency monitoring to get objective data.
03:55
We can perform low-cost mass recruitment for clinical trials,
03:58
and we can make population-scale screening
04:02
feasible for the first time.
04:04
We have the opportunity to start to search
04:05
for the early biomarkers of the disease before it's too late.
04:08
So, taking the first steps towards this today,
04:11
we're launching the Parkinson's Voice Initiative.
04:14
With Aculab and PatientsLikeMe, we're aiming
04:16
to record a very large number of voices worldwide
04:18
to collect enough data to start to tackle these four goals.
04:20
We have local numbers accessible to three quarters
04:23
of a billion people on the planet.
04:25
Anyone healthy or with Parkinson's can call in, cheaply,
04:27
and leave recordings, a few cents each,
04:30
and I'm really happy to announce that we've already hit
04:32
six percent of our target just in eight hours.
04:34
Thank you. (Applause)
04:38
(Applause)
04:41
Tom Rielly: So Max, by taking all these samples of,
04:48
let's say, 10,000 people,
04:51
you'll be able to tell who's healthy and who's not?
04:54
What are you going to get out of those samples?
04:57
Max Little: Yeah. Yeah. So what will happen is that,
04:59
during the call you have to indicate whether or not
05:00
you have the disease or not, you see. TR: Right.
05:02
ML: You see, some people may not do it. They may not get through it.
05:03
But we'll get a very large sample of data that is collected
05:06
from all different circumstances, and it's getting it
05:09
in different circumstances that matter because then
05:12
we are looking at ironing out the confounding factors,
05:14
and looking for the actual markers of the disease.
05:17
TR: So you're 86 percent accurate right now?
05:19
ML: It's much better than that.
05:22
Actually, my student Thanasis, I have to plug him,
05:23
because he's done some fantastic work,
05:25
and now he has proved that it works over the mobile telephone network as well,
05:27
which enables this project, and we're getting 99 percent accuracy.
05:30
TR: Ninety-nine. Well, that's an improvement.
05:34
So what that means is that people will be able to —
05:35
ML: (Laughs)
05:38
TR: People will be able to call in from their mobile phones
05:40
and do this test, and people with Parkinson's could call in,
05:41
record their voice, and then their doctor can check up
05:44
on their progress, see where they're doing in this course of the disease.
05:47
ML: Absolutely.
05:50
TR: Thanks so much. Max Little, everybody.
05:51
ML: Thanks, Tom. (Applause)
05:53
Translator:Joseph Geni
Reviewer:Morton Bast

sponsored links

Max Little - Applied mathematician
Max Little is a mathematician whose research includes a breakthrough technique to monitor – and potentially screen for – Parkinson's disease through simple voice recordings.

Why you should listen

Max Little is an applied mathematician whose goal is to "see connections between subjects, not boundaries … to see how things are related, not how they are different." He has a background in applied mathematics, statistics, signal processing and computational engineering, and his work has been applied across disciplines like biomedicine, extreme rainfall analysis and forecasting, biophysical signal processing, and hydrogeomorphology and open channel flow measurement. Little is best known for his work on the Parkinson's Voice Initiative, in which he and his team developed a cheap and simple tool that uses precise voice analysis software to detect Parkinson's with 99 percent accuracy. Little is a TEDGlobal 2012 Fellow and a Wellcome Trust-MIT Postdoctoral Research Fellow.

The original video is available on TED.com
sponsored links

If you need translations, you can install "Google Translate" extension into your Chrome Browser.
Furthermore, you can change playback rate by installing "Video Speed Controller" extension.

Data provided by TED.

This website is owned and operated by Tokyo English Network.
The developer's blog is here.