TEDxBeaconStreet
Rosalind Picard: An AI smartwatch that detects seizures
羅莎琳皮卡德: 能夠偵測癲癇的人工智慧手錶
Filmed:
Readability: 4.2
1,872,363 views
每年，全世界有超過五萬看似健康其實患有癲癇的人猝死——這就是所謂的癲癇猝死症。人工智慧研究者羅莎琳皮卡德說，其實其中有很大一部分死亡是可以避免的。來聽聽這場演說，了解一下皮卡德怎麼協助開發一種先進的智慧手錶，在癲癇發作之前就先偵測到它，並警告附近的親友來及時協助。
Rosalind Picard - Scientist, inventor, entrepreneur
MIT Professor Rosalind Picard invents technologies that help people better understand emotion and behaviors that impact human health. Full bio
MIT Professor Rosalind Picard invents technologies that help people better understand emotion and behaviors that impact human health. Full bio
Double-click the English transcript below to play the video.
00:13
This is Henry,
0
1747
1487
這位是亨利，
00:15
a cute boy,
1
3258
1806
一位可愛的男孩，
00:17
and when Henry was three,
2
5088
2138
亨利三歲時，
00:19
his mom found him having
some febrile seizures.
some febrile seizures.
3
7250
4935
他的媽媽發現他有點熱痙攣。
00:25
Febrile seizures are seizures that occur
when you also have a fever,
when you also have a fever,
4
13080
4960
熱痙攣就是在痙攣的
同時還加上發燒，
同時還加上發燒，
00:30
and the doctor said,
5
18064
1898
而醫生說：
00:31
"Don't worry too much.
Kids usually outgrow these."
Kids usually outgrow these."
6
19986
2742
「不用太擔心。通常孩子
長大之後就沒有了。」
長大之後就沒有了。」
00:35
When he was four,
he had a convulsive seizure,
he had a convulsive seizure,
7
23506
3981
他四歲時出現了痙攣性癲癇，
00:39
the kind that you lose
consciousness and shake --
consciousness and shake --
8
27511
3262
會失去意識且發抖的那一種——
00:42
a generalized tonic-clonic seizure --
9
30797
2896
全身強直陣攣型發作——
00:45
and while the diagnosis of epilepsy
was in the mail,
was in the mail,
10
33717
6984
當癲癇診斷書還在郵寄的途中，
00:53
Henry's mom went to get him
out of bed one morning,
out of bed one morning,
11
41763
2676
一天早上亨利的媽媽去叫他起床，
00:57
and as she went in his room,
12
45324
1904
她進入他的房間，
01:00
she found his cold, lifeless body.
13
48753
4690
她發現他的身體冰冷，
沒有生命跡象。
沒有生命跡象。
01:07
Henry died of SUDEP,
14
55904
2056
亨利死於 SUDEP，
01:09
sudden unexpected death in epilepsy.
15
57984
2405
即癲癇猝死症。
01:13
I'm curious how many of you
have heard of SUDEP.
have heard of SUDEP.
16
61714
3245
我很好奇，在座有多少人
聽過癲癇猝死症？
聽過癲癇猝死症？
01:17
This is a very well-educated audience,
and I see only a few hands.
and I see only a few hands.
17
65603
4366
各位是教育程度很高的觀眾，
但我只看到幾個人舉手。
但我只看到幾個人舉手。
01:21
SUDEP is when an otherwise
healthy person with epilepsy
healthy person with epilepsy
18
69993
3966
癲癇猝死症指的是
健康的癲癇患者
健康的癲癇患者
01:25
dies and they can't attribute it
to anything they can find in an autopsy.
to anything they can find in an autopsy.
19
73983
3890
在死亡解剖後無法發現任何死因。
01:32
There is a SUDEP
every seven to nine minutes.
every seven to nine minutes.
20
80082
3603
每七到九分鐘就有一個人
發生癲癇猝死症。
發生癲癇猝死症。
01:36
That's on average two per TED Talk.
21
84217
2879
平均每場 TED 演說就有兩個人。
01:43
Now, a normal brain
has electrical activity.
has electrical activity.
22
91411
3659
正常人的大腦都會有電訊活動。
01:47
You can see some of the electrical waves
23
95094
2135
各位可以在這張大腦的圖片上
01:49
coming out of this picture
of a brain here.
of a brain here.
24
97253
3064
看到一些電波。
01:52
And these should look
like typical electrical activity
like typical electrical activity
25
100341
3746
這些是腦電圖從表面讀到的
01:56
that an EEG could read on the surface.
26
104111
2135
典型腦電活動。
01:58
When you have a seizure,
it's a bit of unusual electrical activity,
it's a bit of unusual electrical activity,
27
106270
4417
當你發作時，就會有
不尋常的腦電活動，
不尋常的腦電活動，
02:02
and it can be focal.
28
110711
1385
這可能就是病灶所在。
02:04
It can take place
in just a small part of your brain.
in just a small part of your brain.
29
112120
2500
它可能只發生在大腦的一塊小區域。
02:06
When that happens,
you might have a strange sensation.
you might have a strange sensation.
30
114644
3808
發生時，你可能會有
很奇怪的感覺。
很奇怪的感覺。
02:10
Several could be happening
here in the audience right now,
here in the audience right now,
31
118476
3033
在座當中可能就有幾個人
現在就有這個狀況，
現在就有這個狀況，
02:13
and the person next to you
might not even know.
might not even know.
32
121533
2602
而你旁邊的人可能完全不會知道。
02:16
However, if you have a seizure
where that little brush fire spreads
where that little brush fire spreads
33
124159
3920
然而，如果你腦中的發作狀況
是從小小的火苗散播到
是從小小的火苗散播到
02:20
like a forest fire over the brain,
34
128103
1992
像是森林大火，
02:22
then it generalizes,
35
130119
1674
它就會變成全身性，
02:23
and that generalized seizure
takes your consciousness away
takes your consciousness away
36
131817
4571
而全身性的發作會奪去你的意識，
02:28
and causes you to convulse.
37
136412
1576
造成你抽搐。
02:30
There are more SUDEPs
in the United States every year
in the United States every year
38
138680
4142
在美國，每年癲癇猝死的人數
02:34
than sudden infant death syndrome.
39
142846
2659
高於嬰兒猝死症。
02:38
Now, how many of you have heard
of sudden infant death syndrome?
of sudden infant death syndrome?
40
146298
3002
在座有多少人聽過嬰兒猝死症？
02:41
Right? Pretty much every hand goes up.
41
149324
1956
是吧？幾乎人人都舉手了。
02:43
So what's going on here?
42
151304
1723
這是怎麼回事？
02:45
Why is this so much more common
and yet people haven't heard of it?
and yet people haven't heard of it?
43
153051
5143
為什麼這種疾病這麼普遍，
然而大家卻沒聽說過它呢？
然而大家卻沒聽說過它呢？
02:50
And what can you do to prevent it?
44
158218
2436
還有，要如何預防它呢？
02:52
Well, there are two things,
scientifically shown,
scientifically shown,
45
160678
2302
嗯，科學指出，有兩個方法
02:55
that prevent or reduce the risk of SUDEP.
46
163004
3373
可以預防或減低
癲癇猝死症的風險。
癲癇猝死症的風險。
02:59
The first is: "Follow
your doctor's instructions,
your doctor's instructions,
47
167115
2555
第一：「遵循醫生指示，
03:01
take your medications."
48
169694
1436
按時吃藥。」
03:03
Two-thirds of people who have epilepsy
49
171154
2201
三分之二的癲癇患者
03:05
get it under control
with their medications.
with their medications.
50
173379
2119
都能用藥物來控制。
03:07
The second thing that reduces
the risk of SUDEP is companionship.
the risk of SUDEP is companionship.
51
175848
4171
能減低癲癇猝死症風險的
第二個方法，就是陪伴。
第二個方法，就是陪伴。
03:12
It's having somebody there
at the time that you have a seizure.
at the time that you have a seizure.
52
180043
4694
也就是當你發作的時候，
要有人在你身邊。
要有人在你身邊。
03:16
Now, SUDEP, even though
most of you have never heard of it,
most of you have never heard of it,
53
184761
3921
雖然大部分人沒聽過癲癇猝死症，
03:20
is actually the number two cause
of years of potential life lost
of years of potential life lost
54
188706
5189
但它其實是所有神經疾病當中，
最有可能致死的第二名。
03:25
of all neurological disorders.
55
193919
2976
03:29
The vertical axis is the number of deaths
56
197197
4286
縱軸是死亡人數，
03:33
times the remaining life span,
57
201507
3166
乘上剩下的生命期長，
03:36
so higher is much worse impact.
58
204697
3439
所以越高表示越糟糕的影響。
03:40
SUDEP, however, unlike these others,
59
208522
2974
然而，癲癇猝死症不像其他疾病，
03:43
is something that people right here
could do something to push that down.
could do something to push that down.
60
211520
5716
這裡的人都能夠
做點什麼來壓制住它。
做點什麼來壓制住它。
03:49
Now, what is Roz Picard, an AI researcher,
doing here telling you about SUDEP, right?
doing here telling you about SUDEP, right?
61
217778
7000
為什麼人工智慧研究者
羅莎琳皮卡德
羅莎琳皮卡德
要在這裡跟大家談癲癇猝死症？
03:56
I'm not a neurologist.
62
224802
2521
我不是神經病學家。
03:59
When I was working at the Media Lab
on measurement of emotion,
on measurement of emotion,
63
227347
4215
我在麻省理工媒體實驗室
研究情緒測量時，
研究情緒測量時，
04:03
trying to make our machines
more intelligent about our emotions,
more intelligent about our emotions,
64
231586
3222
我試圖要讓我們的機器
有更高的智能來判讀我們的情緒，
有更高的智能來判讀我們的情緒，
04:06
we started doing a lot of work
measuring stress.
measuring stress.
65
234832
2531
我們開始努力去測量壓力。
04:11
We built lots of sensors
66
239212
1936
我們建造了許多感測器，
04:13
that measured it
in lots of different ways.
in lots of different ways.
67
241172
2381
用許多不同的方式來測量壓力。
04:15
But one of them in particular
68
243577
2320
但，其中有一種特別方式，
04:17
grew out of some of this very old work
with measuring sweaty palms
with measuring sweaty palms
69
245921
4112
出自用電訊號來測量掌心流汗的
04:22
with an electrical signal.
70
250057
1744
古老研究。
04:23
This is a signal of skin conductance
71
251825
1749
這是皮膚電導的訊號，
04:25
that's known to go up
when you get nervous,
when you get nervous,
72
253598
2118
已知當你緊張時，這種訊號會升高，
04:27
but it turns out it also goes up with
a lot of other interesting conditions.
a lot of other interesting conditions.
73
255740
3572
但，結果發現，還有許多其他
有趣的條件也會讓它上升。
有趣的條件也會讓它上升。
04:31
But measuring it with wires on your hand
is really inconvenient.
is really inconvenient.
74
259336
3022
但，手上連結電線來測量
實在不太方便。
實在不太方便。
所以，我們在麻省理工
媒體實驗室裡
媒體實驗室裡
04:34
So we invented a bunch of other ways
of doing this at the MIT Media Lab.
of doing this at the MIT Media Lab.
75
262382
3745
發明了許多其他方法來做這件事。
04:38
And with these wearables,
76
266151
2173
有了這些可穿戴的配備，
04:40
we started to collect the first-ever
clinical quality data 24-7.
clinical quality data 24-7.
77
268348
4805
我們開始能取得一週七天、
一天二十四小時的
一天二十四小時的
高品質臨床資料，
這是前所未有的。
這是前所未有的。
04:45
Here's a picture of what that looked like
78
273609
2533
這張圖上的是
04:48
the first time an MIT student collected
skin conductance on the wrist 24-7.
skin conductance on the wrist 24-7.
79
276166
6595
第一次從一位麻省理工學院
學生的手腕收集到
學生的手腕收集到
一週七天、一天二十四小時的
皮膚電導資料。
皮膚電導資料。
04:55
Let's zoom in a little bit here.
80
283187
2358
咱們把它放大一點。
04:57
What you see is 24 hours
from left to right,
from left to right,
81
285569
3025
從左到右是二十四小時，
05:00
and here is two days of data.
82
288618
1927
圖上的是兩天份的資料。
05:02
And first, what surprised us
83
290569
2682
首先，讓我們驚訝的是，
05:05
was sleep was the biggest
peak of the day.
peak of the day.
84
293275
3346
一天的高峰發生在睡眠時間。
05:08
Now, that sounds broken, right?
85
296645
1750
聽起來是故障了，對吧？
05:10
You're calm when you're asleep,
so what's going on here?
so what's going on here?
86
298419
4178
你沉睡時是很冷靜的，
所以這是怎麼回事？
所以這是怎麼回事？
05:14
Well, it turns out
that our physiology during sleep
that our physiology during sleep
87
302621
2453
結果發現，在睡眠期間的生理機能
05:17
is very different
than our physiology during wake,
than our physiology during wake,
88
305098
2643
和清醒期間非常不同，
05:19
and while there's still a bit of a mystery
89
307765
2184
雖然對於為什麼一天之中的高峰
05:21
why these peaks are usually
the biggest of the day during sleep,
the biggest of the day during sleep,
90
309973
3744
通常出現在睡眠期間，
這仍然有點難以理解，
這仍然有點難以理解，
05:25
we now believe they're related
to memory consolidation
to memory consolidation
91
313741
3170
但我們現在相信，它們和記憶穩固
05:28
and memory formation during sleep.
92
316935
2055
及睡眠期間的記憶形成有關。
05:31
We also saw things
that were exactly what we expected.
that were exactly what we expected.
93
319895
3571
也有一些發現，
完全在我們的預期之中。
完全在我們的預期之中。
05:35
When an MIT student
is working hard in the lab
is working hard in the lab
94
323490
2350
當麻省理工學院學生
在實驗室中努力或在做功課時，
在實驗室中努力或在做功課時，
05:37
or on homeworks,
95
325864
1389
05:39
there is not only emotional stress,
but there's cognitive load,
but there's cognitive load,
96
327277
4547
不僅有情緒壓力，還有認知負荷，
05:43
and it turns out that cognitive load,
cognitive effort, mental engagement,
cognitive effort, mental engagement,
97
331848
4861
結果發現，認知負荷、
認知努力、心理參與、
認知努力、心理參與、
05:48
excitement about learning something --
98
336733
2194
對學習某件事物感到興奮——
05:50
those things also make the signal go up.
99
338951
2337
上述這些也都會造成訊號上升。
05:55
Unfortunately, to the embarrassment
of we MIT professors,
of we MIT professors,
100
343014
3738
不幸的是，讓我們麻省理工學院
教授們很丟臉的是，
教授們很丟臉的是，
05:58
(Laughter)
101
346776
1079
（笑聲）
05:59
the low point every day
is classroom activity.
is classroom activity.
102
347879
4468
一天的低點出現在教室活動期間。
06:04
Now, I am just showing you
one person's data here,
one person's data here,
103
352810
2516
各位在這裡看到的，
只是一個人的資料，
只是一個人的資料，
06:07
but this, unfortunately,
is true in general.
is true in general.
104
355350
2610
但，不幸的是，一般來說
大家都有這個現象。
大家都有這個現象。
06:12
This sweatband has inside it
a homebuilt skin-conductance sensor,
a homebuilt skin-conductance sensor,
105
360400
4389
這條防汗帶裡面
有自製的皮膚電導感測器，
有自製的皮膚電導感測器，
06:16
and one day, one of our undergrads
knocked on my door
knocked on my door
106
364813
5086
有一天，我們的一位
大學生來敲我的門，
大學生來敲我的門，
06:21
right at the end of the December semester,
107
369923
2675
那時是十二月學期末，
06:24
and he said, "Professor Picard,
108
372622
2190
他說：「皮卡德教授，
06:26
can I please borrow
one of your wristband sensors?
one of your wristband sensors?
109
374836
2714
我能不能跟你借一個腕帶感測器？
06:29
My little brother has autism,
he can't talk,
he can't talk,
110
377574
3729
我弟弟有自閉症，他無法說話，
06:33
and I want to see
what's stressing him out."
what's stressing him out."
111
381327
2167
我想了解是什麼讓他很焦慮。」
06:36
And I said, "Sure, in fact,
don't just take one, take two,"
don't just take one, take two,"
112
384434
2858
我說：「當然，
事實上，拿兩條去吧。」
事實上，拿兩條去吧。」
06:39
because they broke easily back then.
113
387316
2792
因為在那時，
這種感測器很容易壞。
這種感測器很容易壞。
06:42
So he took them home,
he put them on his little brother.
he put them on his little brother.
114
390132
2620
他把感測器帶回家，
讓他弟弟戴上。
讓他弟弟戴上。
06:44
Now, I was back in MIT,
looking at the data on my laptop,
looking at the data on my laptop,
115
392776
3169
而我在麻省理工學院裡，
用我的筆記型電腦看資料，
用我的筆記型電腦看資料，
06:47
and the first day, I thought,
"Hmm, that's odd,
"Hmm, that's odd,
116
395969
2861
第一天，我心想：「嗯，真怪，
06:50
he put them on both wrists
instead of waiting for one to break.
instead of waiting for one to break.
117
398854
2991
他把兩個感測器戴在兩隻手腕上，
而不是等一個壞了。
而不是等一個壞了。
06:53
OK, fine, don't follow my instructions."
118
401869
2320
好，不照我的指示就算了。」
06:56
I'm glad he didn't.
119
404213
1718
我很高興他沒照我的指示。
06:58
Second day -- chill.
Looked like classroom activity.
Looked like classroom activity.
120
406377
3206
第二天——冷淡。
看起來就像教室活動期間。
看起來就像教室活動期間。
07:01
(Laughter)
121
409607
1262
（笑聲）
07:03
A few more days ahead.
122
411389
1329
還要觀察幾天。
07:05
The next day, one wrist signal was flat
123
413404
4514
隔天，有一隻手腕的訊號很平坦，
07:09
and the other had
the biggest peak I've ever seen,
the biggest peak I've ever seen,
124
417942
3958
另一隻則有我見過最高的高峰，
07:14
and I thought, "What's going on?
125
422496
2318
我心想：「這是怎麼回事？
07:16
We've stressed people out at MIT
every way imaginable.
every way imaginable.
126
424838
3460
我們在麻省理工學院裡用過
各種想得出來的方式製造焦慮。
各種想得出來的方式製造焦慮。
07:20
I've never seen a peak this big."
127
428933
1738
我從沒見過這麼高的高峰。」
07:24
And it was only on one side.
128
432409
1969
且只發生在一隻手。
07:26
How can you be stressed on one side
of your body and not the other?
of your body and not the other?
129
434402
3144
怎麼可能身體的一邊
很焦慮，另一邊卻不會？
很焦慮，另一邊卻不會？
所以我認為，至少
其中一個感測器壞了。
其中一個感測器壞了。
07:29
So I thought one or both sensors
must be broken.
must be broken.
130
437570
2870
07:33
Now, I'm an electroengineer by training,
131
441185
1917
我是受過訓練的電機工程師，
07:35
so I started a whole bunch of stuff
to try to debug this,
to try to debug this,
132
443126
2669
所以我開始動手去偵錯，
07:37
and long story short,
I could not reproduce this.
I could not reproduce this.
133
445819
2499
長話短說，我無法複製這個狀況。
07:40
So I resorted to old-fashioned debugging.
134
448962
3033
所以，我回去用老式的偵錯方法。
07:44
I called the student at home on vacation.
135
452019
3282
這位學生放假在家，
我打電話給他：
我打電話給他：
07:47
"Hi, how's your little brother?
How's your Christmas?
How's your Christmas?
136
455325
4343
「嗨，你弟弟如何？
聖誕節過得如何？
聖誕節過得如何？
07:51
Hey, do you have any idea
what happened to him?"
what happened to him?"
137
459692
3281
嘿，你知道他發生了什麼事嗎？」
07:54
And I gave this particular date and time,
138
462997
1956
我提出了這個特定的日期、
時間，以及資料。
時間，以及資料。
07:56
and the data.
139
464977
1301
07:58
And he said, "I don't know,
I'll check the diary."
I'll check the diary."
140
466302
3415
他說：「我不知道，
我來查一下日記。」
我來查一下日記。」
08:02
Diary? An MIT student keeps a diary?
141
470850
2908
日記？麻省理工學院的
學生會寫日記？
學生會寫日記？
08:05
So I waited and he came back.
142
473782
1819
我等待著，他回來了。
08:07
He had the exact date and time,
143
475625
1539
他有精確的日期和時間，
08:09
and he says, "That was right before
he had a grand mal seizure."
he had a grand mal seizure."
144
477188
3853
他說：「那剛好就是他
癲癇大發作之前。」
癲癇大發作之前。」
08:14
Now, at the time, I didn't know
anything about epilepsy,
anything about epilepsy,
145
482942
3349
當時，我對癲癇一無所知，
08:18
and did a bunch of research,
146
486315
2159
於是我做了一堆研究，
08:20
realized that another student's dad
is chief of neurosurgery
is chief of neurosurgery
147
488498
3126
我發現有一個學生的爸爸是
波士頓兒童醫院的神經外科主任，
08:23
at Children's Hospital Boston,
148
491648
2000
08:25
screwed up my courage
and called Dr. Joe Madsen.
and called Dr. Joe Madsen.
149
493672
2289
我鼓起勇氣，打電話給
喬馬德森醫生。
喬馬德森醫生。
08:27
"Hi, Dr. Madsen,
my name's Rosalind Picard.
my name's Rosalind Picard.
150
495985
2469
「嗨，馬德森醫生，
我是羅莎琳皮卡德。
我是羅莎琳皮卡德。
08:30
Is it possible somebody could have
151
498478
3896
有沒有可能，有人會發生
08:34
a huge sympathetic
nervous system surge" --
nervous system surge" --
152
502398
4403
交感神經系統突然高漲」——
08:38
that's what drives the skin conductance --
153
506825
2578
皮膚電導就是這麼來的——
08:41
"20 minutes before a seizure?"
154
509427
1672
「然後二十分鐘後就發作？」
08:44
And he says, "Probably not."
155
512172
2119
他說：「應該不會。」
08:47
He says, "It's interesting.
156
515997
1334
他說：「這很有趣。
我們見過病人一隻手臂上的毛聳立，
08:49
We've had people whose hair
stands on end on one arm
stands on end on one arm
157
517355
2768
08:52
20 minutes before a seizure."
158
520147
2080
然後二十分鐘後就發作。」
08:55
And I'm like, "On one arm?"
159
523060
1315
我說：「一隻手臂？」
一開始，我還不想告訴他這件事，
一開始，我還不想告訴他這件事，
08:56
I didn't want to tell him that, initially,
160
524399
2130
08:58
because I thought this was too ridiculous.
161
526553
2389
因為我覺得太荒謬了。
09:00
He explained how this could
happen in the brain,
happen in the brain,
162
528966
2293
他解釋了大腦中
可能發生了什麼事，
可能發生了什麼事，
09:03
and he got interested.
I showed him the data.
I showed him the data.
163
531283
2173
他很感興趣，我把資料給他看。
09:05
We made a whole bunch more devices,
got them safety certified.
got them safety certified.
164
533480
2948
我們又做了一大堆裝置，
且為它們取得安全認證。
且為它們取得安全認證。
09:08
90 families were being
enrolled in a study,
enrolled in a study,
165
536452
2074
一項研究招募了九十個家庭參加，
09:10
all with children who were going
to be monitored 24-7
to be monitored 24-7
166
538550
3229
他們的孩子會受到一週七天，
一天二十四小時的監控，
一天二十四小時的監控，
09:13
with gold-standard EEG on their scalp
167
541803
3371
頭皮會接上高品質的腦波儀，
來讀取大腦活動數據，
來讀取大腦活動數據，
09:17
for reading the brain activity,
168
545198
1543
09:18
video to watch the behavior,
169
546765
1943
有影片可觀察行為，
09:20
electrocardiogram -- ECG --
and now EDA, electrodermal activity,
and now EDA, electrodermal activity,
170
548732
3660
還有心電圖—— ECG ——
現在是看 EDA，膚電活動，
現在是看 EDA，膚電活動，
09:24
to see if there was
something in this periphery
something in this periphery
171
552416
2585
用來尋找周圍是否有什麼我們可以
09:27
that we could easily pick up,
related to a seizure.
related to a seizure.
172
555025
2446
很容易取得和發作有關的訊號。
09:30
We found, in 100 percent
of the first batch of grand mal seizures,
of the first batch of grand mal seizures,
173
558263
6441
我們發現，
第一批的癲癇大發作
全部都伴隨這種超大的膚電反應。
09:36
this whopper of responses
in the skin conductance.
in the skin conductance.
174
564728
3633
09:40
The blue in the middle, the boy's sleep,
175
568385
2003
中間的藍色區段是男孩的
睡眠期間，通常都是一天的高峰。
睡眠期間，通常都是一天的高峰。
09:42
is usually the biggest peak of the day.
176
570412
1972
09:44
These three seizures you see here
are popping out of the forest
are popping out of the forest
177
572408
3632
現在看到的這三次發作特別突出，
09:48
like redwood trees.
178
576064
1476
就像森林中的巨杉。
09:51
Furthermore, when you couple
the skin conductance at the top
the skin conductance at the top
179
579096
3580
此外，如果把上面的膚電資料
09:54
with the movement from the wrist
180
582700
3517
搭配來自手腕的活動，
09:58
and you get lots of data
and train machine learning and AI on it,
and train machine learning and AI on it,
181
586241
4800
就會得到很多資料，可以用來
訓練機器學習和人工智慧，
訓練機器學習和人工智慧，
10:03
you can build an automated AI
that detects these patterns
that detects these patterns
182
591065
4793
可以建造自動化人工智慧，
來偵測這些模式，
來偵測這些模式，
10:07
much better than just
a shake detector can do.
a shake detector can do.
183
595882
4215
比晃動偵測器能做的要多太多。
10:12
So we realized that we needed
to get this out,
to get this out,
184
600121
4267
我們知道我們得要
把這些成果傳出去，
把這些成果傳出去，
10:16
and with the PhD work of Ming-Zher Poh
185
604412
2358
加上傅明哲的博士論文研究
10:18
and later great improvements by Empatica,
186
606794
3027
和 Empatica 公司後續做的大改良，
10:21
this has made progress and the seizure
detection is much more accurate.
detection is much more accurate.
187
609845
3844
讓我們有了進展，
發作偵測的精確度大大提升。
發作偵測的精確度大大提升。
10:25
But we also learned some other things
about SUDEP during this.
about SUDEP during this.
188
613713
2905
過程中，我們也更進一步
了解了癲癇猝死症。
了解了癲癇猝死症。
10:28
One thing we learned is that SUDEP,
189
616642
3412
關於癲癇猝死症，
我們學到的一件事就是，
我們學到的一件事就是，
10:32
while it's rare after
a generalized tonic-clonic seizure,
a generalized tonic-clonic seizure,
190
620078
2912
雖然很少會在全身
強直陣攣型發作之後，
強直陣攣型發作之後，
10:35
that's when it's most likely
to happen -- after that type.
to happen -- after that type.
191
623014
3237
那是它最可能發生的時段——
在那種發作之後。
在那種發作之後。
10:38
And when it happens,
it doesn't happen during the seizure,
it doesn't happen during the seizure,
192
626275
3119
當它發生時，並不會
在發作的過程中發生，
在發作的過程中發生，
10:41
and it doesn't usually happen
immediately afterwards,
immediately afterwards,
193
629418
2866
通常不會緊接著馬上發生，
10:44
but immediately afterwards,
194
632308
1715
但，緊接著，
10:46
when the person just seems
very still and quiet,
very still and quiet,
195
634047
3062
當病人還看似非常安穩的時候，
10:49
they may go into another phase,
where the breathing stops,
where the breathing stops,
196
637133
4947
可能就會進入另一個階段，
在這個階段中，呼吸停止，
在這個階段中，呼吸停止，
10:54
and then after the breathing stops,
later the heart stops.
later the heart stops.
197
642104
3024
在呼吸停止後，心臟接著停止。
10:57
So there's some time
to get somebody there.
to get somebody there.
198
645152
2043
所以還要花些時間才會到那一步。
11:00
We also learned that there is a region
deep in the brain called the amygdala,
deep in the brain called the amygdala,
199
648349
4904
我們也發現，在大腦深處
有一個區域叫做杏仁核，
有一個區域叫做杏仁核，
11:05
which we had been studying
in our emotion research a lot.
in our emotion research a lot.
200
653277
2936
情緒研究常常會去探究這個區域。
11:08
We have two amygdalas,
201
656237
1281
我們有兩個杏仁核，
如果去刺激右邊的那個，
如果去刺激右邊的那個，
11:09
and if you stimulate the right one,
202
657542
1673
11:11
you get a big right
skin conductance response.
skin conductance response.
203
659239
2739
產生的右側膚電反應就會很大。
11:14
Now, you have to sign up right now
for a craniotomy to get this done,
for a craniotomy to get this done,
204
662002
4245
要願意接受顱骨切開術
才能做到這件事。
才能做到這件事。
11:18
not exactly something
we're going to volunteer to do,
we're going to volunteer to do,
205
666271
2833
我們不會想要自願去做，
11:21
but it causes a big right skin
conductance response.
conductance response.
206
669128
2513
但它會造成很大的右側膚電反應。
11:23
Stimulate the left one, big left
skin conductance response on the palm.
skin conductance response on the palm.
207
671665
3853
刺激左邊的，手掌會有
很大的左側膚電反應。
很大的左側膚電反應。
11:27
And furthermore, when somebody
stimulates your amygdala
stimulates your amygdala
208
675542
4040
此外，當有人刺激你的杏仁核，
11:31
while you're sitting there
and you might just be working,
and you might just be working,
209
679606
3419
當時你坐在那裡，可能正在工作，
11:35
you don't show any signs of distress,
210
683049
1888
你不會顯露出任何苦難的徵兆，
11:37
but you stop breathing,
211
685620
1286
但你會停止呼吸，
11:40
and you don't start again
until somebody stimulates you.
until somebody stimulates you.
212
688470
3317
而且除非有人去刺激你，
你不會再次呼吸。
你不會再次呼吸。
11:44
"Hey, Roz, are you there?"
213
692599
1513
「嘿，羅莎，你在嗎？」
11:46
And you open your mouth to talk.
214
694136
1931
你打開你的嘴巴說話。
11:48
As you take that breath to speak,
215
696868
2194
當你吸入空氣以便說話，
11:51
you start breathing again.
216
699086
1505
你又開始呼吸了。
11:54
So we had started with work on stress,
217
702762
3620
所以，我們開始研究壓力，
11:58
which had enabled us
to build lots of sensors
to build lots of sensors
218
706406
2253
這讓我們能夠打造許多感測器，
12:00
that were gathering
high quality enough data
high quality enough data
219
708683
2132
用來收集品質夠高的資料，
12:02
that we could leave the lab
and start to get this in the wild;
and start to get this in the wild;
220
710839
2921
能夠收集實驗室以外的資料；
12:05
accidentally found a whopper
of a response with the seizure,
of a response with the seizure,
221
713784
2864
無意間發現發作帶來的高峰反應，
12:08
neurological activation that can cause
a much bigger response
a much bigger response
222
716672
2859
這種神經系統活化，造成的反應
遠大於傳統的壓力因子；
遠大於傳統的壓力因子；
12:11
than traditional stressors;
223
719555
1310
12:12
lots of partnership with hospitals
and an epilepsy monitoring unit,
and an epilepsy monitoring unit,
224
720889
3198
和許多醫院及一個
癲癇監控單位合作，
癲癇監控單位合作，
12:16
especially Children's Hospital Boston
225
724111
1828
特別是波士頓兒童醫院，
及布萊根婦女醫院；
及布萊根婦女醫院；
12:17
and the Brigham;
226
725963
1151
12:19
and machine learning and AI on top of this
227
727138
2525
此外還有機器學習和人工智慧，
12:21
to take and collect lots more data
228
729687
3021
來取得、收集更多資料，
12:24
in service of trying
to understand these events
to understand these events
229
732732
2629
用來試圖了解這些事件，
12:27
and if we could prevent SUDEP.
230
735385
1594
希望能夠預防癲癇猝死症。
12:30
This is now commercialized by Empatica,
231
738023
3563
現在，Empatica 已經將它商業化，
12:33
a start-up that I had
the privilege to cofound,
the privilege to cofound,
232
741610
2889
它是間新創公司，
我很榮幸是共同創辦人，
我很榮幸是共同創辦人，
12:36
and the team there has done an amazing job
improving the technology
improving the technology
233
744523
4206
那裡的團隊把這項技術
做了很驚人的改良，
做了很驚人的改良，
12:40
to make a very beautiful sensor
234
748753
1759
做出一個很棒的感測器，
12:42
that not only tells time and does steps
and sleep and all that good stuff,
and sleep and all that good stuff,
235
750536
3859
它不僅能夠報時，
也提供睡眠相關各種資訊，
也提供睡眠相關各種資訊，
12:46
but this is running real-time
AI and machine learning
AI and machine learning
236
754419
3096
而且還有即時的
人工智慧和機器學習，
人工智慧和機器學習，
12:49
to detect generalized
tonic-clonic seizures
tonic-clonic seizures
237
757539
2618
可以偵測全身強直陣攣型發作，
12:52
and send an alert for help
238
760181
2151
並在我即將發作且失去意識時
12:54
if I were to have a seizure
and lose consciousness.
and lose consciousness.
239
762356
3192
發出求助警告。
12:57
This just got FDA-approved
240
765572
2816
它剛得到食品及藥物
管理局的核准，
管理局的核准，
13:00
as the first smartwatch
to get approved in neurology.
to get approved in neurology.
241
768412
4140
是神經學領域中第一支
被核准的智慧手錶。
被核准的智慧手錶。
13:05
(Applause)
242
773923
7000
（掌聲）
13:14
Now, the next slide is what made
my skin conductance go up.
my skin conductance go up.
243
782674
3531
下一張投影片呈現的是
我的皮膚電導上升的原因。
我的皮膚電導上升的原因。
13:18
One morning, I'm checking my email
244
786811
1706
有天早上，我在讀電子郵件，
13:20
and I see a story from a mom
245
788541
2297
看到了一位母親的故事，
13:22
who said she was in the shower,
246
790862
1886
她說她當時正在淋浴，
13:24
and her phone was
on the counter by the shower,
on the counter by the shower,
247
792772
2453
她的手機在淋浴間的檯子上，
13:27
and it said her daughter
might need her help.
might need her help.
248
795249
2157
手機說她的女兒
可能需要她的協助。
可能需要她的協助。
13:30
So she interrupts her shower and goes
running to her daughter's bedroom,
running to her daughter's bedroom,
249
798196
3381
所以她淋浴到一半
就衝到她女兒的臥房，
就衝到她女兒的臥房，
她發現她的女兒面朝下
趴在床上，發青，沒有呼吸。
趴在床上，發青，沒有呼吸。
13:33
and she finds her daughter
facedown in bed, blue and not breathing.
facedown in bed, blue and not breathing.
250
801601
3743
13:37
She flips her over -- human stimulation --
251
805368
3281
她把她翻過來——人類刺激——
13:40
and her daughter takes a breath,
and another breath,
and another breath,
252
808673
3683
她的女兒吸了一口氣，
接著再一口，
接著再一口，
13:44
and her daughter turns pink and is fine.
253
812380
3954
她的女兒漸漸恢復血色，沒事了。
13:49
I think I turned white reading this email.
254
817551
3224
我想我在讀這封信時
我應該是臉色發白的。
我應該是臉色發白的。
13:52
My first response is,
"Oh no, it's not perfect.
"Oh no, it's not perfect.
255
820799
2322
我的第一個反應是：
「喔，不，它不夠完美。
「喔，不，它不夠完美。
13:55
The Bluetooth could break,
the battery could die.
the battery could die.
256
823145
2325
藍芽可能會斷掉，電池可能會用完。
這些都可能會出錯。
不要仰賴這個裝置。
不要仰賴這個裝置。
13:57
All these things could go wrong.
Don't rely on this."
Don't rely on this."
257
825494
2477
她說：「沒關係，我知道
沒有任何技術是完美的。
沒有任何技術是完美的。
13:59
And she said, "It's OK.
I know no technology is perfect.
I know no technology is perfect.
258
827995
3408
14:03
None of us can always
be there all the time.
be there all the time.
259
831427
2163
沒有人能隨時都在身邊。
14:06
But this, this device plus AI
260
834844
4342
但，這個裝置加上人工智慧，
14:11
enabled me to get there in time
to save my daughter's life."
to save my daughter's life."
261
839210
3104
讓我能及時趕到，
救了我女兒一命。」
救了我女兒一命。」
14:17
Now, I've been mentioning children,
262
845933
1906
我剛剛一直在談孩子，
14:19
but SUDEP peaks, actually,
among people in their 20s, 30s and 40s,
among people in their 20s, 30s and 40s,
263
847863
5707
但其實，癲癇猝死症最常發生在
二十、三十、四十多歲的人身上，
二十、三十、四十多歲的人身上，
14:25
and the next line I'm going to put up
264
853594
1816
我接下來要說的，
14:27
is probably going to make
some people uncomfortable,
some people uncomfortable,
265
855434
2453
可能會讓一些人不舒服，
14:29
but it's less uncomfortable
than we'll all be
than we'll all be
266
857911
2374
但，如果這份名單
包括到你認識的人，
包括到你認識的人，
14:32
if this list is extended
to somebody you know.
to somebody you know.
267
860309
3070
就會更不舒服。
14:36
Could this happen to somebody you know?
268
864437
2338
這有沒有可能發生在
你認識的人身上？
你認識的人身上？
14:39
And the reason I bring up
this uncomfortable question
this uncomfortable question
269
867197
2742
我之所以提出這個不舒服的問題，
14:41
is because one in 26 of you
will have epilepsy at some point,
will have epilepsy at some point,
270
869963
4912
是因為二十六個人當中就有
一個人，會在某個時點罹患癲癇，
一個人，會在某個時點罹患癲癇，
14:46
and from what I've been learning,
271
874899
2235
據我所知，
14:49
people with epilepsy often don't tell
their friends and their neighbors
their friends and their neighbors
272
877158
3334
罹患癲癇的人通常
不會告訴朋友和鄰居。
不會告訴朋友和鄰居。
14:52
that they have it.
273
880516
1158
所以，如果你願意讓他們
使用人工智慧之類的東西
使用人工智慧之類的東西
14:53
So if you're willing to let them
use an AI or whatever
use an AI or whatever
274
881698
4921
14:58
to summon you in a moment
of possible need,
of possible need,
275
886643
4414
在他們可能需要協助時呼叫你，
15:03
if you would let them know that,
276
891081
1618
如果你能讓他們知道這點，
15:04
you could make a difference in their life.
277
892723
2354
你就能讓他們的人生有所不同。
15:07
Why do all this hard work to build AIs?
278
895728
2851
為什麼要花這麼多功夫
來打造人工智慧？
來打造人工智慧？
15:12
A couple of reasons here:
279
900101
1201
這裡有幾個理由：
15:13
one is Natasha, the girl who lived,
280
901326
2163
第一個理由是娜塔莎，
活下來的那位女孩，
活下來的那位女孩，
15:15
and her family wanted me
to tell you her name.
to tell you her name.
281
903513
2392
她的家人要我把
她的名字告訴大家。
她的名字告訴大家。
15:18
Another is her family
282
906977
1612
另一個理由是她的家人
15:20
and the wonderful people out there
283
908613
2191
以及其他美好的人，
15:22
who want to be there to support people
who have conditions
who have conditions
284
910828
2825
他們想要支援那些
過去不敢告訴他人
過去不敢告訴他人
15:25
that they've felt uncomfortable
in the past mentioning to others.
in the past mentioning to others.
285
913677
3293
自己患有某些疾病的人。
15:29
And the other reason is all of you,
286
917901
1873
還有一個理由是你們所有人，
15:31
because we have the opportunity
to shape the future of AI.
to shape the future of AI.
287
919798
5332
因為我們有機會形塑
人工智慧的未來。
人工智慧的未來。
15:37
We can actually change it,
288
925154
2357
我們能夠改變它，
15:39
because we are the ones building it.
289
927535
2404
因為我們就是建造它的人。
15:41
So let's build AI
290
929963
1977
所以，咱們來打造出
15:43
that makes everybody's lives better.
291
931964
3292
能改善大家生活的人工智慧。
15:47
Thank you.
292
935982
1151
謝謝。
15:49
(Applause)
293
937157
4414
（掌聲）
ABOUT THE SPEAKER
Rosalind Picard - Scientist, inventor, entrepreneurMIT Professor Rosalind Picard invents technologies that help people better understand emotion and behaviors that impact human health.
Why you should listen
Rosalind Picard wrote Affective Computing, a book that outlines how to give machines the skills of emotional intelligence, credited for launching the growing field of affective computing. Named one of seven "Tech SuperHeros to Watch" by CNN, Picard cofounded two companies that have commercialized technologies invented by her team at MIT: Empatica, which created the first AI-based smartwatch cleared by FDA in Neurology, and Affectiva, which provides emotion AI software. Picard teaches and leads research in affective computing at the MIT Media Lab in Cambridge; works as MIT's faculty chair for MindHandHeart, a campus-wide wellbeing initiative; and serves as chief scientist for Empatica, creating AI-based analytics and wearables to improve human wellbeing.
Rosalind Picard | Speaker | TED.com