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

114,963 views ・ 2012-08-07

TED


Please double-click on the English subtitles below to play the video.

00:00
Translator: Joseph Geni Reviewer: Morton Bast
0
0
7000
00:15
So, well, I do applied math,
1
15927
2103
00:18
and this is a peculiar problem
2
18030
1524
00:19
for anyone who does applied math, is that
3
19554
2173
00:21
we are like management consultants.
4
21727
1933
00:23
No one knows what the hell we do.
5
23660
1946
00:25
So I am going to give you some -- attempt today
6
25606
2274
00:27
to try and explain to you what I do.
7
27880
2293
00:30
So, dancing is one of the most human of activities.
8
30173
3321
00:33
We delight at ballet virtuosos and tap dancers
9
33494
3682
00:37
you will see later on.
10
37176
1148
00:38
Now, ballet requires an extraordinary level of expertise
11
38324
2690
00:41
and a high level of skill,
12
41014
2914
00:43
and probably a level of initial suitability
13
43928
2531
00:46
that may well have a genetic component to it.
14
46459
1847
00:48
Now, sadly, neurological disorders such as Parkinson's disease
15
48306
3393
00:51
gradually destroy this extraordinary ability,
16
51699
2087
00:53
as it is doing to my friend Jan Stripling, who was
17
53786
2323
00:56
a virtuoso ballet dancer in his time.
18
56109
2967
00:59
So great progress and treatment has been made over the years.
19
59076
3054
01:02
However, there are 6.3 million people worldwide
20
62130
2944
01:05
who have the disease, and they have to live with
21
65074
3448
01:08
incurable weakness, tremor, rigidity
22
68522
2568
01:11
and the other symptoms that go along with the disease,
23
71090
1857
01:12
so what we need are objective tools
24
72947
2383
01:15
to detect the disease before it's too late.
25
75330
3057
01:18
We need to be able to measure progression objectively,
26
78387
2554
01:20
and ultimately, the only way we're going to know
27
80941
3173
01:24
when we actually have a cure is when we have
28
84114
2192
01:26
an objective measure that can answer that for sure.
29
86306
3398
01:29
But frustratingly, with Parkinson's disease
30
89704
2850
01:32
and other movement disorders, there are no biomarkers,
31
92554
2353
01:34
so there's no simple blood test that you can do,
32
94907
2229
01:37
and the best that we have is like
33
97136
1802
01:38
this 20-minute neurologist test.
34
98938
2241
01:41
You have to go to the clinic to do it. It's very, very costly,
35
101179
2458
01:43
and that means that, outside the clinical trials,
36
103637
2757
01:46
it's just never done. It's never done.
37
106394
2728
01:49
But what if patients could do this test at home?
38
109122
3077
01:52
Now, that would actually save on a difficult trip to the clinic,
39
112199
2098
01:54
and what if patients could do that test themselves, right?
40
114297
4254
01:58
No expensive staff time required.
41
118551
1920
02:00
Takes about $300, by the way,
42
120471
1418
02:01
in the neurologist's clinic to do it.
43
121889
1993
02:03
So what I want to propose to you as an unconventional way
44
123882
2681
02:06
in which we can try to achieve this,
45
126563
1514
02:08
because, you see, in one sense, at least,
46
128077
1808
02:09
we are all virtuosos like my friend Jan Stripling.
47
129885
3256
02:13
So here we have a video of the vibrating vocal folds.
48
133141
3755
02:16
Now, this is healthy and this is somebody making speech sounds,
49
136896
3229
02:20
and we can think of ourselves as vocal ballet dancers,
50
140125
3464
02:23
because we have to coordinate all of these vocal organs
51
143589
2214
02:25
when we make sounds, and we all actually
52
145803
2295
02:28
have the genes for it. FoxP2, for example.
53
148098
2296
02:30
And like ballet, it takes an extraordinary level of training.
54
150394
2713
02:33
I mean, just think how long it takes a child to learn to speak.
55
153107
2585
02:35
From the sound, we can actually track
56
155692
2382
02:38
the vocal fold position as it vibrates,
57
158074
2281
02:40
and just as the limbs are affected in Parkinson's,
58
160355
2543
02:42
so too are the vocal organs.
59
162898
2781
02:45
So on the bottom trace, you can see an example of
60
165679
1880
02:47
irregular vocal fold tremor.
61
167559
1698
02:49
We see all the same symptoms.
62
169257
1168
02:50
We see vocal tremor, weakness and rigidity.
63
170425
2930
02:53
The speech actually becomes quieter and more breathy
64
173355
2104
02:55
after a while, and that's one of the example symptoms of it.
65
175459
2233
02:57
So these vocal effects can actually be quite subtle,
66
177692
2847
03:00
in some cases, but with any digital microphone,
67
180539
3216
03:03
and using precision voice analysis software
68
183755
2545
03:06
in combination with the latest in machine learning,
69
186300
2409
03:08
which is very advanced by now,
70
188709
1578
03:10
we can now quantify exactly where somebody lies
71
190287
2886
03:13
on a continuum between health and disease
72
193173
2881
03:16
using voice signals alone.
73
196054
2596
03:18
So these voice-based tests, how do they stack up against
74
198650
2314
03:20
expert clinical tests? We'll, they're both non-invasive.
75
200964
2150
03:23
The neurologist's test is non-invasive. They both use existing infrastructure.
76
203114
3982
03:27
You don't have to design a whole new set of hospitals to do it.
77
207096
3004
03:30
And they're both accurate. Okay, but in addition,
78
210100
2302
03:32
voice-based tests are non-expert.
79
212402
3327
03:35
That means they can be self-administered.
80
215729
1992
03:37
They're high-speed, take about 30 seconds at most.
81
217721
2580
03:40
They're ultra-low cost, and we all know what happens.
82
220301
2294
03:42
When something becomes ultra-low cost,
83
222595
2440
03:45
it becomes massively scalable.
84
225035
2296
03:47
So here are some amazing goals that I think we can deal with now.
85
227331
3675
03:51
We can reduce logistical difficulties with patients.
86
231006
2426
03:53
No need to go to the clinic for a routine checkup.
87
233432
2312
03:55
We can do high-frequency monitoring to get objective data.
88
235744
2320
03:58
We can perform low-cost mass recruitment for clinical trials,
89
238064
4105
04:02
and we can make population-scale screening
90
242169
2115
04:04
feasible for the first time.
91
244284
1596
04:05
We have the opportunity to start to search
92
245880
2202
04:08
for the early biomarkers of the disease before it's too late.
93
248082
3541
04:11
So, taking the first steps towards this today,
94
251623
2758
04:14
we're launching the Parkinson's Voice Initiative.
95
254381
2126
04:16
With Aculab and PatientsLikeMe, we're aiming
96
256507
2232
04:18
to record a very large number of voices worldwide
97
258739
1928
04:20
to collect enough data to start to tackle these four goals.
98
260667
3140
04:23
We have local numbers accessible to three quarters
99
263807
1700
04:25
of a billion people on the planet.
100
265507
1610
04:27
Anyone healthy or with Parkinson's can call in, cheaply,
101
267117
3077
04:30
and leave recordings, a few cents each,
102
270194
2139
04:32
and I'm really happy to announce that we've already hit
103
272333
2190
04:34
six percent of our target just in eight hours.
104
274523
3543
04:38
Thank you. (Applause)
105
278066
3751
04:41
(Applause)
106
281817
6320
04:48
Tom Rielly: So Max, by taking all these samples of,
107
288137
3575
04:51
let's say, 10,000 people,
108
291712
2776
04:54
you'll be able to tell who's healthy and who's not?
109
294488
2854
04:57
What are you going to get out of those samples?
110
297342
1685
04:59
Max Little: Yeah. Yeah. So what will happen is that,
111
299027
1830
05:00
during the call you have to indicate whether or not
112
300857
1657
05:02
you have the disease or not, you see. TR: Right.
113
302514
1267
05:03
ML: You see, some people may not do it. They may not get through it.
114
303781
2507
05:06
But we'll get a very large sample of data that is collected
115
306288
2717
05:09
from all different circumstances, and it's getting it
116
309005
3408
05:12
in different circumstances that matter because then
117
312413
1905
05:14
we are looking at ironing out the confounding factors,
118
314318
3384
05:17
and looking for the actual markers of the disease.
119
317702
2161
05:19
TR: So you're 86 percent accurate right now?
120
319863
2497
05:22
ML: It's much better than that.
121
322360
1194
05:23
Actually, my student Thanasis, I have to plug him,
122
323554
1720
05:25
because he's done some fantastic work,
123
325274
1870
05:27
and now he has proved that it works over the mobile telephone network as well,
124
327144
3770
05:30
which enables this project, and we're getting 99 percent accuracy.
125
330914
3390
05:34
TR: Ninety-nine. Well, that's an improvement.
126
334304
1576
05:35
So what that means is that people will be able to —
127
335880
2201
05:38
ML: (Laughs)
128
338081
1852
05:39
TR: People will be able to call in from their mobile phones
129
339933
1906
05:41
and do this test, and people with Parkinson's could call in,
130
341839
3072
05:44
record their voice, and then their doctor can check up
131
344911
2870
05:47
on their progress, see where they're doing in this course of the disease.
132
347781
2681
05:50
ML: Absolutely.
133
350462
970
05:51
TR: Thanks so much. Max Little, everybody.
134
351432
1743
05:53
ML: Thanks, Tom. (Applause)
135
353175
5157
About this website

This site will introduce you to YouTube videos that are useful for learning English. You will see English lessons taught by top-notch teachers from around the world. Double-click on the English subtitles displayed on each video page to play the video from there. The subtitles scroll in sync with the video playback. If you have any comments or requests, please contact us using this contact form.

https://forms.gle/WvT1wiN1qDtmnspy7