請雙擊下方英文字幕播放視頻。
譯者: 易帆 余
審譯者: Adrienne Lin
00:12
So you go to the doctor
and get some tests.
0
12811
3321
你去看醫生,接受了一些檢查。
00:16
The doctor determines
that you have high cholesterol
1
16674
2620
醫生診斷出你的膽固醇過高,
00:19
and you would benefit
from medication to treat it.
2
19318
3171
建議你服藥治療可能有幫助。
00:22
So you get a pillbox.
3
22981
1556
所以,你拿到了藥罐子。
00:25
You have some confidence,
4
25505
1199
你有點信心,
00:26
your physician has some confidence
that this is going to work.
5
26728
2937
你的醫師也有信心,認為這藥會有效。
00:29
The company that invented it did
a lot of studies, submitted it to the FDA.
6
29689
3553
發明這個藥的公司做了很多的研究,
然後呈送給食品藥物管理局。
00:33
They studied it very carefully,
skeptically, they approved it.
7
33266
3107
他們很仔細、審慎地研究,
並核准了這藥物上市。
00:36
They have a rough idea of how it works,
8
36397
1889
他們大概知道這藥物如何運作,
00:38
they have a rough idea
of what the side effects are.
9
38310
2453
也大略知道會有什麼副作用,
00:40
It should be OK.
10
40787
1150
應該沒問題。
00:42
You have a little more
of a conversation with your physician
11
42864
2818
你跟醫師又多聊了一會,
00:45
and the physician is a little worried
because you've been blue,
12
45706
2963
而醫師有點擔心,因為你很憂鬱,
00:48
haven't felt like yourself,
13
48693
1293
精神欠佳。
00:50
you haven't been able to enjoy things
in life quite as much as you usually do.
14
50010
3731
無法像平常一樣盡情享受生活點滴。
00:53
Your physician says, "You know,
I think you have some depression.
15
53765
3186
你的醫師說:
「我認為你有一點精神憂鬱,
00:57
I'm going to have to give
you another pill."
16
57792
2315
我再開個藥給你。」
01:00
So now we're talking
about two medications.
17
60934
2483
所以,我們現在有兩種藥了。
01:03
This pill also -- millions
of people have taken it,
18
63441
3104
這個藥也有好幾百萬人服用過,
01:06
the company did studies,
the FDA looked at it -- all good.
19
66569
3631
公司做了研究,食品藥物管理局
也檢查過,全部都沒問題。
01:10
Think things should go OK.
20
70823
2057
想一下,這東西沒問題,OK的。
01:12
Think things should go OK.
21
72904
2197
想一下,這東西沒問題,OK的。
01:15
Well, wait a minute.
22
75125
1439
但,請等一下。
01:16
How much have we studied
these two together?
23
76588
3517
我們對這兩種藥混在一起吃
做了多少研究?
01:20
Well, it's very hard to do that.
24
80630
2300
其實,這很難評估。
01:22
In fact, it's not traditionally done.
25
82954
2130
事實上,傳統上都不會做。
01:25
We totally depend on what we call
"post-marketing surveillance,"
26
85108
5518
在藥物上市後,我們完全倚賴一種
叫做「上市後監察系統」的機制,
01:30
after the drugs hit the market.
27
90650
1880
01:32
How can we figure out
if bad things are happening
28
92996
2848
我們要如何確認,兩種藥之間
01:35
between two medications?
29
95868
1357
是否有什麼不好的事會發生?
01:37
Three? Five? Seven?
30
97249
2030
三種?五種?七種呢?
01:39
Ask your favorite person
who has several diagnoses
31
99708
2415
問你身邊有各種疾病在身的人,
01:42
how many medications they're on.
32
102147
1834
他們正在吃多少藥。
01:44
Why do I care about this problem?
33
104530
1580
為什麼我在乎這個問題?
01:46
I care about it deeply.
34
106134
1157
我非常在乎。
01:47
I'm an informatics and data science guy
and really, in my opinion,
35
107315
4304
我是念資訊和數據科學的人,
真的,在我看來,
01:51
the only hope -- only hope --
to understand these interactions
36
111643
3745
了解藥物彼此間的交互影響
唯一的希望只有
01:55
is to leverage lots
of different sources of data
37
115412
3056
運用不同來源的龐大資料,
01:58
in order to figure out
when drugs can be used together safely
38
118492
3556
才能找出這些藥
何時可以安全地一起服用,
02:02
and when it's not so safe.
39
122072
1777
以及何時不行。
02:04
So let me tell you a data science story.
40
124615
2051
所以,讓我來告訴各位
一個數據科學的故事。
02:06
And it begins with my student Nick.
41
126690
2154
這要從我的學生尼克開始講起。
02:08
Let's call him "Nick,"
because that's his name.
42
128868
2380
我們就稱呼他為尼克吧,
因為那就是他的本名。
02:11
(Laughter)
43
131272
1592
(笑聲)
02:12
Nick was a young student.
44
132888
1201
尼克很年輕,
02:14
I said, "You know, Nick, we have
to understand how drugs work
45
134113
3079
我說:「尼克,
我們必須了解藥物如何運作,
02:17
and how they work together
and how they work separately,
46
137216
2626
以及藥物在一起會如何運作、
分開會如何運作,
02:19
and we don't have a great understanding.
47
139866
1922
而我們並沒有了解很深。」
02:21
But the FDA has made available
an amazing database.
48
141812
2405
但食品藥物管理局已經
有一個很驚人的資料庫,
02:24
It's a database of adverse events.
49
144241
1699
是一個藥物不良反應通報資料庫。
02:26
They literally put on the web --
50
146321
1642
資料真的直接放在網路上
02:27
publicly available, you could all
download it right now --
51
147987
3119
供大眾查詢,你現在就可以全部下載,
02:31
hundreds of thousands
of adverse event reports
52
151130
3627
從病人、醫生、公司、藥劑師通報上來
02:34
from patients, doctors,
companies, pharmacists.
53
154781
3760
好幾百萬個的藥物不良反應通報。
02:38
And these reports are pretty simple:
54
158565
1749
這些報告都相當簡單:
02:40
it has all the diseases
that the patient has,
55
160338
2658
上面有病人所有疾病
02:43
all the drugs that they're on,
56
163020
1767
及所有藥物的使用狀況,
02:44
and all the adverse events,
or side effects, that they experience.
57
164811
3818
還有他們經歷過的
所有不良反應事件或副作用。
02:48
It is not all of the adverse events
that are occurring in America today,
58
168653
3436
雖然沒有現今在美國
發生的所有不良反應事件,
02:52
but it's hundreds and hundreds
of thousands of drugs.
59
172113
2578
但卻有上百萬種藥物資科。
02:54
So I said to Nick,
60
174715
1299
所以,我跟尼克說:
02:56
"Let's think about glucose.
61
176038
1826
「我們來想一想葡萄糖。
02:57
Glucose is very important,
and we know it's involved with diabetes.
62
177888
3567
葡萄糖非常重要,而且
大家都知道它與糖尿病有關。
03:01
Let's see if we can understand
glucose response.
63
181479
3970
讓我們來看看是否可以
了解葡萄糖的反應。」
03:05
I sent Nick off. Nick came back.
64
185473
2458
我請尼克去找資料,
03:08
"Russ," he said,
65
188248
1786
他回來後說:「洛斯,
03:10
"I've created a classifier that can
look at the side effects of a drug
66
190351
5112
我已經建造了一個分辨器,
可以透過這個資料庫
03:15
based on looking at this database,
67
195487
2051
來檢視一種藥物的副作用,
03:17
and can tell you whether that drug
is likely to change glucose or not."
68
197562
4271
而且還可以告訴你,這個藥
會否改變病人血糖狀況。」
03:21
He did it. It was very simple, in a way.
69
201857
2016
他用一個方法做到了,很簡單。
03:23
He took all the drugs
that were known to change glucose
70
203897
2635
他把所有已知會改變葡萄糖的藥物
03:26
and a bunch of drugs
that don't change glucose,
71
206556
2389
及所有不會改變的藥物拿出來做比較,
03:28
and said, "What's the difference
in their side effects?
72
208969
2888
「它們之間的副作用有什麼分別?
03:31
Differences in fatigue? In appetite?
In urination habits?"
73
211881
4852
疲勞狀況上的差異?食慾上的差異?
排尿習慣上的差異?」
03:36
All those things conspired
to give him a really good predictor.
74
216757
2960
所有這些事情都可以協助他
做出一個很棒的預測器。
03:39
He said, "Russ, I can predict
with 93 percent accuracy
75
219741
2548
他說:「洛斯,我能預測
哪種藥可改變血糖,
03:42
when a drug will change glucose."
76
222313
1572
準確率可以高達93%。」
03:43
I said, "Nick, that's great."
77
223909
1416
我說:「尼克,這太棒了!」
他是個年輕的學生,
03:45
He's a young student,
you have to build his confidence.
78
225349
2896
你必須建立他的信心。
03:48
"But Nick, there's a problem.
79
228269
1390
「但,尼克,有一個問題。
03:49
It's that every physician in the world
knows all the drugs that change glucose,
80
229683
3960
就是全世界的醫師都知道
這些藥會改變葡萄糖,
03:53
because it's core to our practice.
81
233667
2038
因為這是我們實務上的核心。
03:55
So it's great, good job,
but not really that interesting,
82
235729
3722
所以,你很棒,幹得好,
但並沒有人對這有興趣,
03:59
definitely not publishable."
83
239475
1531
絕對還不適合公布你的研究結果。」
04:01
(Laughter)
84
241030
1014
(笑聲 )
04:02
He said, "I know, Russ.
I thought you might say that."
85
242068
2550
他說:「我知道,洛斯。
我知道你可能會這麼說。」
04:04
Nick is smart.
86
244642
1152
尼克很聰明。
04:06
"I thought you might say that,
so I did one other experiment.
87
246149
2874
「我知道你會這麼說,
所以我多做了另一項實驗。
04:09
I looked at people in this database
who were on two drugs,
88
249047
2928
我仔細觀察資料庫裡
同時服用兩種藥的人,
04:11
and I looked for signals similar,
glucose-changing signals,
89
251999
4422
然後尋找他們之間
葡萄糖改變的相似訊號,
04:16
for people taking two drugs,
90
256445
1624
04:18
where each drug alone
did not change glucose,
91
258093
5569
但前提是,這些藥單獨服用
不會改變葡萄糖,
04:23
but together I saw a strong signal."
92
263686
2460
一起服用時,會有強烈訊號的藥物。」
04:26
And I said, "Oh! You're clever.
Good idea. Show me the list."
93
266170
3149
我說:「喔!你真聰明,
好主意,讓我看一下清單。」
04:29
And there's a bunch of drugs,
not very exciting.
94
269343
2254
有一大堆藥,並沒有令人非常興奮。
04:31
But what caught my eye
was, on the list there were two drugs:
95
271621
3932
但引起我注意的是,清單上有兩種藥:
04:35
paroxetine, or Paxil, an antidepressant;
96
275577
3393
帕羅西汀或稱克憂果,
這是一種治療憂鬱症的藥,
04:39
and pravastatin, or Pravachol,
a cholesterol medication.
97
279756
3570
還有普伐他汀或稱美百樂,
一種治療心臟疾病的藥。
04:43
And I said, "Huh. There are millions
of Americans on those two drugs."
98
283936
4283
然後我說:「哈!有上百萬
美國人正在服用這兩種藥」。
04:48
In fact, we learned later,
99
288243
1246
事實上,我們之後才知道,
04:49
15 million Americans on paroxetine
at the time, 15 million on pravastatin,
100
289513
6032
當時有1500萬美國人正在服用帕羅西汀,
1500萬人正在服用普伐他汀,
04:55
and a million, we estimated, on both.
101
295569
2817
而我們預估有100萬人,
同時服用這兩個藥。
04:58
So that's a million people
102
298767
1254
所以,有100萬人
05:00
who might be having some problems
with their glucose
103
300045
2453
可能有葡萄糖上的問題,
05:02
if this machine-learning mumbo jumbo
that he did in the FDA database
104
302522
3206
如果他用食品藥物管理局的資料庫
05:05
actually holds up.
105
305752
1254
做的機械學習判讀器真的有用的話。
05:07
But I said, "It's still not publishable,
106
307030
1927
但我說:「還是不能發表,
05:08
because I love what you did
with the mumbo jumbo,
107
308981
2296
因為我雖然喜歡你做的
05:11
with the machine learning,
108
311301
1246
機械學習判讀器,
05:12
but it's not really standard-of-proof
evidence that we have."
109
312571
3864
但我們沒有真正的證明標準
來證明我們是正確的。」
05:17
So we have to do something else.
110
317618
1589
所以,我們來必須做些其他事來驗證。
05:19
Let's go into the Stanford
electronic medical record.
111
319231
2876
我們去找史丹佛的電子病例紀錄。
05:22
We have a copy of it
that's OK for research,
112
322131
2064
我們有一個副本,可以用來研究,
05:24
we removed identifying information.
113
324219
2046
我們移除了病人個資。
05:26
And I said, "Let's see if people
on these two drugs
114
326581
2503
我說:「讓我們來看看,
服用這兩種藥的人
05:29
have problems with their glucose."
115
329108
1769
是否有葡萄糖上的疾病。」
05:31
Now there are thousands
and thousands of people
116
331242
2207
在史丹佛病例紀錄中有成千上萬的人
05:33
in the Stanford medical records
that take paroxetine and pravastatin.
117
333473
3459
同時服用這兩種藥。
05:36
But we needed special patients.
118
336956
1799
但我們需要特定病患。
05:38
We needed patients who were on one of them
and had a glucose measurement,
119
338779
4597
我們需要已經做葡萄糖檢測
且服用其中一種藥的病人,
05:43
then got the second one and had
another glucose measurement,
120
343400
3449
另外再找到另一個已經做過
另一個葡萄糖檢測的病人,
05:46
all within a reasonable period of time --
something like two months.
121
346873
3615
全部都在合理期間做的,
例如兩個月內。
05:50
And when we did that,
we found 10 patients.
122
350512
3159
當我們開始著手進行時,
我們找到十個病人。
05:54
However, eight out of the 10
had a bump in their glucose
123
354592
4538
然而,十個人裡面
有八個葡萄糖異常增加現象,
05:59
when they got the second P --
we call this P and P --
124
359154
2645
在他們服用第二個P時
─我們稱呼這個叫 P&P─
06:01
when they got the second P.
125
361823
1310
當他們服用了第二個 P。
06:03
Either one could be first,
the second one comes up,
126
363157
2562
哪一個先服用都行,
當第二個藥服用後,
06:05
glucose went up
20 milligrams per deciliter.
127
365743
2847
葡萄糖濃度每公升會增加20毫克。
06:08
Just as a reminder,
128
368614
1158
提醒各位一下,
06:09
you walk around normally,
if you're not diabetic,
129
369796
2325
如果你能正常走動,沒有糖尿病,
06:12
with a glucose of around 90.
130
372145
1359
你的葡萄糖濃度約90毫克/公升。
06:13
And if it gets up to 120, 125,
131
373528
2076
如果上升到120、125,
06:15
your doctor begins to think
about a potential diagnosis of diabetes.
132
375628
3450
你的醫生會開始認為
你有潛在的糖尿病症狀。
06:19
So a 20 bump -- pretty significant.
133
379102
2991
所以,一下子增加20是相當明顯的。
06:22
I said, "Nick, this is very cool.
134
382601
1904
我說:「尼克,這很酷。
06:25
But, I'm sorry, we still
don't have a paper,
135
385616
2053
但,很抱歉,我們仍然沒辦法寫報告,
06:27
because this is 10 patients
and -- give me a break --
136
387693
2579
因為只有十個病人,饒了我吧,
06:30
it's not enough patients."
137
390296
1245
病人樣本數根本不夠。」
06:31
So we said, what can we do?
138
391565
1306
所以,那怎麼辦?
06:32
And we said, let's call our friends
at Harvard and Vanderbilt,
139
392895
2976
我們來打電話給哈佛
及范德堡大學的朋友,
06:35
who also -- Harvard in Boston,
Vanderbilt in Nashville,
140
395895
2587
就是波士頓的哈佛
及納許維爾的范德堡,
06:38
who also have electronic
medical records similar to ours.
141
398506
2821
他們都有跟我們很像的
電子病歷紀錄。
06:41
Let's see if they can find
similar patients
142
401351
2020
讓我們看看,他們是否
也可以找到相同的病人,
06:43
with the one P, the other P,
the glucose measurements
143
403395
3276
也有我們需要的已經服用這兩種藥,
06:46
in that range that we need.
144
406695
1600
並做過葡萄糖檢測的病人。
06:48
God bless them, Vanderbilt
in one week found 40 such patients,
145
408787
4955
上天保佑,范德堡一個星期內找到40個
06:53
same trend.
146
413766
1189
有同樣趨勢的病人。
06:55
Harvard found 100 patients, same trend.
147
415804
3620
哈佛找到100個有同樣趨勢的病人。
06:59
So at the end, we had 150 patients
from three diverse medical centers
148
419448
4281
所以,最後,我們從三個不同的
醫學中心找到150個病人
07:03
that were telling us that patients
getting these two drugs
149
423753
3297
服用過這兩種藥,
然後有葡萄糖異常增加現象。
07:07
were having their glucose bump
somewhat significantly.
150
427074
2703
07:10
More interestingly,
we had left out diabetics,
151
430317
2810
有趣的是,我們沒有考慮糖尿病患者,
07:13
because diabetics already
have messed up glucose.
152
433151
2317
因為糖尿病患者本身的
血糖濃度就已經很混亂。
07:15
When we looked
at the glucose of diabetics,
153
435492
2238
當我們觀察糖尿病患者的血糖濃度時,
07:17
it was going up 60 milligrams
per deciliter, not just 20.
154
437754
3435
會上升到每公升60毫克,
不只20毫克。
07:21
This was a big deal, and we said,
"We've got to publish this."
155
441760
3452
這事情很重要,我們說:
「我們必須發佈這件事。」
07:25
We submitted the paper.
156
445236
1179
我們遞交報告,
07:26
It was all data evidence,
157
446439
2111
裡面全部都是資料證明,
07:28
data from the FDA, data from Stanford,
158
448574
2483
有來自食品藥物管理局、史丹佛的資料、
07:31
data from Vanderbilt, data from Harvard.
159
451081
1946
有來自范德堡、哈佛醫學院的資料,
07:33
We had not done a single real experiment.
160
453051
2396
我們完全沒有做任何實驗。
07:36
But we were nervous.
161
456495
1296
但我們很緊張。
07:38
So Nick, while the paper
was in review, went to the lab.
162
458201
3730
所以,當報告送去審核時,
尼克就去了實驗室。
07:41
We found somebody
who knew about lab stuff.
163
461955
2462
我們找到會做實驗的人。
07:44
I don't do that.
164
464441
1393
我不做實驗的。
07:45
I take care of patients,
but I don't do pipettes.
165
465858
2417
我會看病人,但我不會做分量管。
07:49
They taught us how to feed mice drugs.
166
469420
3053
他們教我們如何餵老鼠吃藥。
07:52
We took mice and we gave them
one P, paroxetine.
167
472864
2414
我們給第一組老鼠餵食帕羅西汀,
07:55
We gave some other mice pravastatin.
168
475302
2508
給第二組老鼠餵食普伐他汀。
07:57
And we gave a third group
of mice both of them.
169
477834
3595
第三組的老鼠兩種藥都餵食。
08:01
And lo and behold, glucose went up
20 to 60 milligrams per deciliter
170
481893
3946
驚奇的是,葡萄糖每公升上升20到60毫克,
老鼠也有相同的反應。
08:05
in the mice.
171
485863
1171
08:07
So the paper was accepted
based on the informatics evidence alone,
172
487058
3158
所以,只有資料證據的報告被接受了,
但我們在最後加了註記說,
08:10
but we added a little note at the end,
173
490240
1894
08:12
saying, oh by the way,
if you give these to mice, it goes up.
174
492158
2899
如果把藥物給老鼠,葡萄糖也會上升。
08:15
That was great, and the story
could have ended there.
175
495081
2508
太棒了,故事其實就到這裡結束。
08:17
But I still have six and a half minutes.
176
497613
1997
但,我還有六分半鐘。
08:19
(Laughter)
177
499634
2807
(笑聲)
08:22
So we were sitting around
thinking about all of this,
178
502465
2815
所以,我們坐下來想一下所有的事,
08:25
and I don't remember who thought
of it, but somebody said,
179
505304
2735
我忘記誰曾經說過,但有人說:
08:28
"I wonder if patients
who are taking these two drugs
180
508063
3201
「不曉得同時服用這兩種藥的病人,
08:31
are noticing side effects
of hyperglycemia.
181
511288
3553
是否有注意到高血糖症的副作用。
08:34
They could and they should.
182
514865
1496
他們可能知道,也必須知道。
08:36
How would we ever determine that?"
183
516761
1877
我們要如何確定?」
08:39
We said, well, what do you do?
184
519551
1443
我們說,好吧,你會怎麼做?
08:41
You're taking a medication,
one new medication or two,
185
521018
2580
你服用了一種藥,一個或兩個新藥,
08:43
and you get a funny feeling.
186
523622
1538
然後你感覺怪怪的。
08:45
What do you do?
187
525184
1151
你會怎麼做?
08:46
You go to Google
188
526359
1151
你會去問 Google,
08:47
and type in the two drugs you're taking
or the one drug you're taking,
189
527534
3349
然後搜尋你在服用的一或兩個藥名,
08:50
and you type in "side effects."
190
530907
1603
然後加上「副作用」。
08:52
What are you experiencing?
191
532534
1356
你會找到什麼?
08:54
So we said OK,
192
534239
1151
所以,我們說,好,
08:55
let's ask Google if they will share
their search logs with us,
193
535414
3056
我們來問 Google 能否
跟我們分享搜尋紀錄,
08:58
so that we can look at the search logs
194
538494
1833
讓我們可以觀察搜尋紀錄,
09:00
and see if patients are doing
these kinds of searches.
195
540351
2565
看是否有病人也在做同樣的搜尋。
09:02
Google, I am sorry to say,
denied our request.
196
542940
3275
很抱歉我得這麼說,
但 Google 拒絕了我們的請求。
09:06
So I was bummed.
197
546819
1151
所以,我很煩惱。
09:07
I was at a dinner with a colleague
who works at Microsoft Research
198
547994
3166
我跟一個在微軟研究室的同事吃晚餐時,
09:11
and I said, "We wanted to do this study,
199
551184
1941
我跟他說:「我們想做這個研究,
09:13
Google said no, it's kind of a bummer."
200
553149
1880
Google 說不行,我有點煩惱。」
09:15
He said, "Well, we have
the Bing searches."
201
555053
2086
他說:「我們有 Bing 搜尋引擎啊。」
09:18
(Laughter)
202
558195
3483
(笑聲)
09:22
Yeah.
203
562805
1267
是啊!
09:24
That's great.
204
564096
1151
太棒了。
09:25
Now I felt like I was --
205
565271
1151
現在,我感覺...
09:26
(Laughter)
206
566446
1000
(笑聲)
09:27
I felt like I was talking to Nick again.
207
567470
2412
我好像又在鼓勵尼克一樣。
09:30
He works for one of the largest
companies in the world,
208
570437
2624
他在全世界數一數二的公司上班,
09:33
and I'm already trying
to make him feel better.
209
573085
2206
我已經開始要安慰他了。
09:35
But he said, "No, Russ --
you might not understand.
210
575315
2445
但他說:「不,洛斯,你可能沒搞懂。
09:37
We not only have Bing searches,
211
577784
1500
我們不只有 Bing 啊,
09:39
but if you use Internet Explorer
to do searches at Google,
212
579308
3340
如果你用 IE 在 Google、
09:42
Yahoo, Bing, any ...
213
582672
1891
雅虎、Bing 等任何搜尋引擎,
09:44
Then, for 18 months, we keep that data
for research purposes only."
214
584587
3643
之後18個月,我們保留這些數據
僅做研究目的使用。
09:48
I said, "Now you're talking!"
215
588254
1936
我說:「這才像話嘛!」
09:50
This was Eric Horvitz,
my friend at Microsoft.
216
590214
2198
這就是我的微軟朋友艾瑞克.霍維茲。
09:52
So we did a study
217
592436
1695
我們做了一項研究,
09:54
where we defined 50 words
that a regular person might type in
218
594155
4619
我們定義出了50個
如果一般人有高血糖症時
會鍵入的關鍵字,
09:58
if they're having hyperglycemia,
219
598798
1602
10:00
like "fatigue," "loss of appetite,"
"urinating a lot," "peeing a lot" --
220
600424
4762
像是疲勞、沒食慾、頻尿等。
10:05
forgive me, but that's one
of the things you might type in.
221
605210
2767
請原諒我,但這些就是
你可能會鍵入的關鍵字。
10:08
So we had 50 phrases
that we called the "diabetes words."
222
608001
2790
所以,我們有了50個短語,
我們稱之為「糖尿病關鍵字」。
10:10
And we did first a baseline.
223
610815
2063
我們先設定了一條基準線。
10:12
And it turns out
that about .5 to one percent
224
612902
2704
原來,網路上有包含這些關鍵字的搜尋
10:15
of all searches on the Internet
involve one of those words.
225
615630
2982
占了0.5~1%的比例。
10:18
So that's our baseline rate.
226
618636
1742
所以,這就是我們的基準線率,
10:20
If people type in "paroxetine"
or "Paxil" -- those are synonyms --
227
620402
4143
如果大家鍵入「帕羅西汀」或「克憂果」
──這些是同義字──
10:24
and one of those words,
228
624569
1215
以及剛剛其中一個關鍵字,
10:25
the rate goes up to about two percent
of diabetes-type words,
229
625808
4890
那糖尿病類型的基準線率會上升到2%,
10:30
if you already know
that there's that "paroxetine" word.
230
630722
3044
如果你已經知道
「帕羅西汀」這個字的話。
10:34
If it's "pravastatin," the rate goes up
to about three percent from the baseline.
231
634191
4547
如果是「普伐他汀」,
那比率會從基準線率上升到3%。
10:39
If both "paroxetine" and "pravastatin"
are present in the query,
232
639171
4390
如果「帕羅西汀」
和「普伐他汀」同時出現,
10:43
it goes up to 10 percent,
233
643585
1669
那會上升到10%,
10:45
a huge three- to four-fold increase
234
645278
3461
有3到4倍的增加,
10:48
in those searches with the two drugs
that we were interested in,
235
648763
3389
用這兩種藥搜尋,會出現
我們感興趣的字在裡面,
10:52
and diabetes-type words
or hyperglycemia-type words.
236
652176
3566
像是糖尿病類的字
或高血糖症類的字。
我們發佈了這個研究,
10:56
We published this,
237
656216
1265
10:57
and it got some attention.
238
657505
1466
並得到一些關注。
10:58
The reason it deserves attention
239
658995
1778
它值得被關注的原因是,
11:00
is that patients are telling us
their side effects indirectly
240
660797
4312
病人會透過搜尋,
直接告訴我們藥物的副作用。
11:05
through their searches.
241
665133
1156
11:06
We brought this
to the attention of the FDA.
242
666313
2138
我們得到了食品藥物管理局的關注。
11:08
They were interested.
243
668475
1269
他們很感興趣。
11:09
They have set up social media
surveillance programs
244
669768
3606
他們已經成立社會媒體監測計畫,
11:13
to collaborate with Microsoft,
245
673398
1751
與微軟展開合作,
11:15
which had a nice infrastructure
for doing this, and others,
246
675173
2794
他們有良好的設備來做這些事,
11:17
to look at Twitter feeds,
247
677991
1282
可以觀察推特的動態、
11:19
to look at Facebook feeds,
248
679297
1716
觀察臉書的動態、
11:21
to look at search logs,
249
681037
1311
觀察搜尋日誌、
11:22
to try to see early signs that drugs,
either individually or together,
250
682372
4909
嘗試觀察引發問題的
無論單一藥物或混合藥物的早期症狀。
11:27
are causing problems.
251
687305
1589
11:28
What do I take from this?
Why tell this story?
252
688918
2174
我從這件事學到什麼?
為什麼要講這個故事?
11:31
Well, first of all,
253
691116
1207
首先,
11:32
we have now the promise
of big data and medium-sized data
254
692347
4037
我們現在有大數據及中型數據稱腰,
11:36
to help us understand drug interactions
255
696408
2918
來幫助我們了解藥物的相互作用,
11:39
and really, fundamentally, drug actions.
256
699350
2420
以及真實、基本的藥物作用。
11:41
How do drugs work?
257
701794
1413
藥物是如何作用?
11:43
This will create and has created
a new ecosystem
258
703231
2836
這個將會創造一個新的生態系統,
11:46
for understanding how drugs work
and to optimize their use.
259
706091
3267
來幫助我們了解藥物如何運作
以及有效使用它們。
11:50
Nick went on; he's a professor
at Columbia now.
260
710303
2659
尼克繼續往前走,
他現在是哥倫比亞的教授。
11:52
He did this in his PhD
for hundreds of pairs of drugs.
261
712986
4072
他用好幾百對藥物做為博士研究。
11:57
He found several
very important interactions,
262
717082
2517
他找到一些非常重要的藥物交互作用,
11:59
and so we replicated this
263
719623
1214
所以,我們複製這個模式,
12:00
and we showed that this
is a way that really works
264
720861
2574
展示出利用這樣做
12:03
for finding drug-drug interactions.
265
723459
2339
來尋找藥與藥之間的作用真的有效。
12:06
However, there's a couple of things.
266
726282
1734
然而,還有一些事。
12:08
We don't just use pairs
of drugs at a time.
267
728040
3046
我們不會同時一次只服用兩種藥。
12:11
As I said before, there are patients
on three, five, seven, nine drugs.
268
731110
4469
就如我之前所說的,
有病人一次是服用三、五、七、九種藥。
12:15
Have they been studied with respect
to their nine-way interaction?
269
735981
3642
他們有認真研究
這九種藥的相互作用嗎?
12:19
Yes, we can do pair-wise,
A and B, A and C, A and D,
270
739647
4208
沒錯,我們可以做成對的藥,
A+B、A+C、A+D,
12:23
but what about A, B, C,
D, E, F, G all together,
271
743879
4286
但如果同一個病人
同時服用ABCDEFG,
12:28
being taken by the same patient,
272
748189
1762
12:29
perhaps interacting with each other
273
749975
2118
那可能會互相產生那些作用?
12:32
in ways that either makes them
more effective or less effective
274
752117
3778
藥效更好或更不好?
12:35
or causes side effects
that are unexpected?
275
755919
2332
或造成那些意想不到的副作用呢?
12:38
We really have no idea.
276
758275
1827
我們真的不知道。
12:40
It's a blue sky, open field
for us to use data
277
760126
3756
它是個開放式的藍天領域,
讓我們可以使用數據,
12:43
to try to understand
the interaction of drugs.
278
763906
2502
來嘗試了解藥物彼此間的作用。
12:46
Two more lessons:
279
766848
1370
另外兩件事:
12:48
I want you to think about the power
that we were able to generate
280
768242
4199
我想要各位去想想
我們所創造出來的力量,
12:52
with the data from people who had
volunteered their adverse reactions
281
772465
4711
就是我們已經可以透過藥劑師、
病人本身、病人的醫師,
12:57
through their pharmacists,
through themselves, through their doctors,
282
777200
3269
來取得自願者身上
他們的藥物不良反應,
13:00
the people who allowed the databases
at Stanford, Harvard, Vanderbilt,
283
780493
3667
這些人同意他們的資料可以被
史丹佛、哈佛、范德堡醫學院
13:04
to be used for research.
284
784184
1427
來做研究使用。
13:05
People are worried about data.
285
785929
1445
大家都擔心個資問題。
13:07
They're worried about their privacy
and security -- they should be.
286
787398
3187
他們擔心自己的隱私及安全
──他們必須要擔心。
我們需要保全系統。
13:10
We need secure systems.
287
790609
1151
13:11
But we can't have a system
that closes that data off,
288
791784
3406
但我們不能有一個
把資料關起來的系統,
13:15
because it is too rich of a source
289
795214
2752
因為它的資源太豐盛了,
13:17
of inspiration, innovation and discovery
290
797990
3971
它對醫學界的鼓舞、
創新、發現新事物
13:21
for new things in medicine.
291
801985
1578
實在太重要了。
13:24
And the final thing I want to say is,
292
804494
1794
最後,我想說的是,
13:26
in this case we found two drugs
and it was a little bit of a sad story.
293
806312
3357
我們發現這兩個藥的案例,
的確是令人難過的故事。
13:29
The two drugs actually caused problems.
294
809693
1921
這兩個藥一起服用真的會有問題。
13:31
They increased glucose.
295
811638
1475
同時服用會增加葡萄糖,
13:33
They could throw somebody into diabetes
296
813137
2446
會造成一個原本沒糖尿病的人
13:35
who would otherwise not be in diabetes,
297
815607
2294
發生糖尿病情形,
13:37
and so you would want to use
the two drugs very carefully together,
298
817925
3175
所以,各位如果想一起使用
這兩種藥,一定要非常小心,
13:41
perhaps not together,
299
821124
1151
最好不要一起服用,
13:42
make different choices
when you're prescribing.
300
822299
2340
當你要開處方簽時,
看看有沒有不同的選擇。
13:44
But there was another possibility.
301
824663
1846
但,也有其他的可能。
13:46
We could have found
two drugs or three drugs
302
826533
2344
我們或許能找到兩或三種藥,
13:48
that were interacting in a beneficial way.
303
828901
2261
一起服用時也許可以更有效。
13:51
We could have found new effects of drugs
304
831616
2712
我們或許也可以找到
13:54
that neither of them has alone,
305
834352
2160
藥物本身沒有的作用,
13:56
but together, instead
of causing a side effect,
306
836536
2493
但在一起服用時不但沒有產生副作用,
13:59
they could be a new and novel treatment
307
839053
2425
反而產生新作用,有可能變成最新的
14:01
for diseases that don't have treatments
308
841502
1882
絕症疾病治療方式,
14:03
or where the treatments are not effective.
309
843408
2007
或者原本的治療方式完全是無效的。
14:05
If we think about drug treatment today,
310
845439
2395
如果我們想想現今的藥物治療方式,
14:07
all the major breakthroughs --
311
847858
1752
所有的重大突破──
14:09
for HIV, for tuberculosis,
for depression, for diabetes --
312
849634
4297
愛滋病、肺結核、
憂鬱症,糖尿病──
14:13
it's always a cocktail of drugs.
313
853955
2830
總像是藥物雞尾酒。
14:16
And so the upside here,
314
856809
1730
這件事的好處是,
14:18
and the subject for a different
TED Talk on a different day,
315
858563
2849
也許哪一天不同的TED主題,
我們又會來到這裡分享,
14:21
is how can we use the same data sources
316
861436
2593
我們要如何用同樣的資料來源
14:24
to find good effects
of drugs in combination
317
864053
3563
來找到藥物混用時產生的好效果,
14:27
that will provide us new treatments,
318
867640
2175
它將提供我們新的治療方式,
14:29
new insights into how drugs work
319
869839
1852
以及對藥物如何作用提供新的見解,
14:31
and enable us to take care
of our patients even better?
320
871715
3786
並且讓我們的病人得到更好的照顧。
14:35
Thank you very much.
321
875525
1166
非常謝謝各位。
14:36
(Applause)
322
876715
3499
(掌聲)
New videos
關於本網站
本網站將向您介紹對學習英語有用的 YouTube 視頻。 您將看到來自世界各地的一流教師教授的英語課程。 雙擊每個視頻頁面上顯示的英文字幕,從那裡播放視頻。 字幕與視頻播放同步滾動。 如果您有任何意見或要求,請使用此聯繫表與我們聯繫。