So you go to the doctor and get some tests.
你去看医生的时候做了一些检查，
The doctor determines that you have high cholesterol and you would benefit from medication to treat it.
医生说你的血脂高，所以吃药会有帮助。
So you get a pillbox. You have some confidence, your physician has some confidence that this is going to work.
那么你就买了一盒药，你有信心，你的医生也有信心，觉得这会对你有帮助。
The company that invented it did a lot of studies, submitted it to the FDA.
发明这个的药物公司做过很多研究，呈送到FDA。
They studied it very carefully, skeptically, they approved it.
他们充满怀疑地研究，后来许可了。
They have a rough idea of how it works, they have a rough idea of what the side effects are.
他们大概地知道这个药的机理和药物的副作用。
It should be OK. You have a little more of a conversation with your physician
它应该没问题。你跟你的医生又谈了一会儿，
and the physician is a little worried because you've been blue, haven't felt like yourself,
医生对你有点儿担心，因为你有些抑郁，感觉你不是自己，
you haven't been able to enjoy things in life quite as much as you usually do.
你对生活不像以前那么充满兴趣。
Your physician says, "You know, I think you have some depression. I'm going to have to give you another pill."
你的医生说：“你知道吗，我觉得你有些抑郁。我会给你另一种药。”
So now we're talking about two medications.
所以，现在我们所谈论的是两种药物。
This pill also -- millions of people have taken it, the company did studies, the FDA looked at it -- all good.
这种药--上百万人服用过，公司做过很多研究，FDA许可的--不会错。
Think things should go OK. Think things should go OK.
你会想应该没问题，应该没问题。
Well, wait a minute. How much have we studied these two together?
等等，我们知道两种药物一起服用的研究吗？
Well, it's very hard to do that. In fact, it's not traditionally done.
而这是很难做到的。实际上，还从来没有。
We totally depend on what we call "post-marketing surveillance," after the drugs hit the market.
在药物上市之后，我们完全地依赖于我们所谓的“市场后监测”。
How can we figure out if bad things are happening between two medications? Three? Five? Seven?
我们如何能够弄清楚，在两种、三种、五种或者七种药物混合之后，会有哪些坏处呢？
Ask your favorite person who has several diagnoses how many medications they're on.
问问你身边有各种疾病在身的人，他们正在吃多少药。
Why do I care about this problem? I care about it deeply.
为什么我在乎这个问题？我非常在乎。
I'm an informatics and data science guy and really, in my opinion,
我是念信息和数据科学的人，真的，在我看来，
the only hope -- only hope -- to understand these interactions is to leverage lots of different sources of data
了解药物彼此间的交互影响，唯一的希望只有运用不同来源的庞大数据，
in order to figure out when drugs can be used together safely and when it's not so safe.
才能找出这些药何时可以安全地一起服用，以及何时不行。
So let me tell you a data science story. And it begins with my student Nick.
所以，让我来告诉各位一个数据科学的故事。这要从我的学生尼克开始讲起。
Let's call him "Nick," because that's his name.
我们就称呼他为尼克吧，因为那就是他的本名。
Nick was a young student. I said, You know, Nick,
尼克很年轻，我说：尼克，
we have to understand how drugs work and how they work together and how they work separately,
我们必须了解药物如何运作，以及药物在一起会如何运作、分开会如何运作，
and we don't have a great understanding. But the FDA has made available an amazing database.
而我们并没有了解很深。但食品药物管理局已经有一个很惊人的数据库，
It's a database of adverse events.
是一个药物不良反应通报数据库。
They literally put on the web -- publicly available, you could all download it right now
数据真的直接放在网络上供大众查询，你现在就可以全部下载，
hundreds of thousands of adverse event reports from patients, doctors, companies, pharmacists.
从病人、医生、公司、药剂师通报上来好几百万个的药物不良反应通报。
And these reports are pretty simple: it has all the diseases that the patient has,
这些报告都相当简单：上面有病人所有疾病及所有药物的使用状况，
all the drugs that they're on, and all the adverse events, or side effects, that they experience.
还有他们经历过的所有不良反应事件或副作用。
It is not all of the adverse events that are occurring in America today,
虽然没有现今在美国发生的所有不良反应事件，
but it's hundreds and hundreds of thousands of drugs.
但却有上百万种药物资科。
So I said to Nick, Let's think about glucose. Glucose is very important, and we know it's involved with diabetes.
所以，我跟尼克说：我们来想一想葡萄糖。葡萄糖非常重要，而且大家都知道它与糖尿病有关。
Let's see if we can understand glucose response.
让我们来看看是否可以了解葡萄糖的反应。
I sent Nick off. Nick came back. "Russ," he said,
我请尼克去找资料，他回来后说：“洛斯，
"I've created a classifier that can look at the side effects of a drug based on looking at this database,
我已经建造了一个分辨器，可以透过这个数据库来检视一种药物的副作用，
and can tell you whether that drug is likely to change glucose or not."
而且还可以告诉你，这个药会否改变病人血糖状况。”
He did it. It was very simple, in a way.
他用一个方法做到了，很简单。
He took all the drugs that were known to change glucose and a bunch of drugs that don't change glucose, and said,
他把所有已知会改变葡萄糖的药物及所有不会改变的药物拿出来做比较，
"What's the difference in their side effects? Differences in fatigue? In appetite? In urination habits?"
“它们之间的副作用有什么分别？疲劳状况上的差异？食欲上的差异？排尿习惯上的差异？”
All those things conspired to give him a really good predictor.
所有这些事情都可以协助他做出一个很棒的预测器。
He said, "Russ, I can predict with 93 percent accuracy when a drug will change glucose."
他说：“洛斯，我能预测哪种药可改变血糖，准确率可以高达93%。”
I said, "Nick, that's great." He's a young student, you have to build his confidence.
我说：“尼克，这太棒了！”他是个年轻的学生，你必须建立他的信心。
"But Nick, there's a problem.
“但，尼克，有一个问题。
It's that every physician in the world knows all the drugs that change glucose, because it's core to our practice.
就是全世界的医师都知道这些药会改变葡萄糖，因为这是我们实务上的核心。
So it's great, good job, but not really that interesting, definitely not publishable."
所以，你很棒，干得好，但并没有人对这有兴趣，绝对还不适合公布你的研究结果。”
He said, "I know, Russ. I thought you might say that." Nick is smart.
他说：“我知道，洛斯。我知道你可能会这么说。”尼克很聪明。
"I thought you might say that, so I did one other experiment.
“我知道你会这么说，所以我多做了另一项实验。
I looked at people in this database who were on two drugs,
我仔细观察数据库里同时服用两种药的人，
and I looked for signals similar, glucose-changing signals, for people taking two drugs,
然后寻找他们之间葡萄糖改变的相似讯号，但前提是，
where each drug alone did not change glucose, but together I saw a strong signal."
这些药单独服用不会改变葡萄糖，一起服用时，会有强烈讯号的药物。”
And I said, "Oh! You're clever. Good idea. Show me the list."
我说：“喔！你真聪明，好主意，让我看一下清单。”
And there's a bunch of drugs, not very exciting.
有一大堆药，并没有令人非常兴奋。
But what caught my eye was, on the list there were two drugs:
但引起我注意的是，清单上有两种药：
paroxetine, or Paxil, an antidepressant; and pravastatin, or Pravachol, a cholesterol medication.
帕罗西汀或称克忧果，这是一种治疗忧郁症的药；还有普伐他汀或称美百乐，一种治疗心脏疾病的药。
And I said, "Huh. There are millions of Americans on those two drugs."
然后我说：“哈！有上百万美国人正在服用这两种药。”
In fact, we learned later, 15 million Americans on paroxetine at the time,
事实上，我们之后才知道，当时有1500万美国人正在服用帕罗西汀，
15 million on pravastatin, and a million, we estimated, on both.
1500万人正在服用普伐他汀，而我们预估有100万人同时服用这两个药。
So that's a million people who might be having some problems with their glucose
所以，有100万人可能有葡萄糖上的问题，
if this machine-learning mumbo jumbo that he did in the FDA database actually holds up.
如果他用食品药物管理局的数据库做的机械学习判读器真的有用的话。
But I said, "It's still not publishable, because I love what you did with the mumbo jumbo, with the machine learning,
但我说：“还是不能发表，因为我虽然喜欢你做的机械学习判读器，
but it's not really standard-of-proof evidence that we have."
但我们没有真正的证明标准来证明我们是正确的。”
So we have to do something else. Let's go into the Stanford electronic medical record.
所以，我们来必须做些其他事来验证。我们去找史丹佛的电子病例纪录。
We have a copy of it that's OK for research, we removed identifying information.
我们有一个副本，可以用来研究，我们移除了病人个资。
And I said, "Let's see if people on these two drugs have problems with their glucose."
我说：“让我们来看看，服用这两种药的人是否有葡萄糖上的疾病。”
Now there are thousands and thousands of people in the Stanford medical records that take paroxetine and pravastatin.
在斯坦福病例纪录中，有成千上万的人同时服用这两种药。
But we needed special patients. We needed patients who were on one of them and had a glucose measurement,
但我们需要特定病患。我们需要已经做葡萄糖检测且服用其中一种药的病人，
then got the second one and had another glucose measurement, all within a reasonable period of time -- something like two months.
另外再找到另一个已经做过另一个葡萄糖检测的病人，全部都在合理期间做的，例如两个月内。

And when we did that, we found 10 patients.
当我们开始着手进行时，我们找到十个病人。
However, eight out of the 10 had a bump in their glucose
然而，十个人里面有八个葡萄糖异常增加现象，
when they got the second P -- we call this P and P--when they got the second P.
在他们服用第二个P时--我们称呼这个叫P&P--当他们服用了第二个P。
Either one could be first, the second one comes up, glucose went up 20 milligrams per deciliter.
哪一个先服用都行，当第二个药服用后，葡萄糖浓度每公升会增加20毫克。
Just as a reminder, you walk around normally, if you're not diabetic, with a glucose of around 90.
提醒各位一下，如果你能正常走动，没有糖尿病，你的葡萄糖浓度约90毫克/公升。
And if it gets up to 120, 125, your doctor begins to think about a potential diagnosis of diabetes.
如果上升到120、125，你的医生会开始认为你有潜在的糖尿病症状。
So a 20 bump -- pretty significant.
所以，一下子增加20是相当明显的。
I said, "Nick, this is very cool.
我说：“尼克，这很酷。
But, I'm sorry, we still don't have a paper, because this is 10 patients and -- give me a break -- it's not enough patients."
但，很抱歉，我们仍然没办法写报告，因为只有十个病人，饶了我吧，病人样本数根本不够。”
So we said, what can we do? And we said, let's call our friends at Harvard and Vanderbilt,
所以，那怎么办？我们来打电话给哈佛及范德堡大学的朋友，
who also -- Harvard in Boston, Vanderbilt in Nashville, who also have electronic medical records similar to ours.
就是波士顿的哈佛及纳许维尔的范德堡，他们都有跟我们很像的电子病历纪录。
Let's see if they can find similar patients with the one P, the other P, the glucose measurements in that range that we need.
让我们看看，他们是否也可以找到相同的病人，也有我们需要的已经服用这两种药，并做过葡萄糖检测的病人。
God bless them, Vanderbilt in one week found 40 such patients, same trend. Harvard found 100 patients, same trend.
上天保佑，范德堡一个星期内找到40个有同样趋势的病人。哈佛找到100个有同样趋势的病人。
So at the end, we had 150 patients from three diverse medical centers
所以，最后，我们从三个不同的医学中心找到150个病人
that were telling us that patients getting these two drugs were having their glucose bump somewhat significantly.
他们服用过这两种药，然后有葡萄糖异常增加现象。
More interestingly, we had left out diabetics, because diabetics already have messed up glucose.
有趣的是，我们没有考虑糖尿病患者，因为糖尿病患者本身的血糖浓度就已经很混乱。
When we looked at the glucose of diabetics, it was going up 60 milligrams per deciliter, not just 20.
当我们观察糖尿病患者的血糖浓度时，会上升到每公升60毫克，不只20毫克。
This was a big deal, and we said, "We've got to publish this." We submitted the paper.
这事情很重要，我们说：“我们必须发布这件事。”我们递交了报告。
It was all data evidence, data from the FDA, data from Stanford, data from Vanderbilt, data from Harvard.
里面全部都是数据证明，有来自食品药物管理局、史丹佛的数据、有来自范德堡、哈佛医学院的资料。
We had not done a single real experiment.
我们完全没有做任何实验。
But we were nervous. So Nick, while the paper was in review, went to the lab.
但我们很紧张。所以，当报告送去审核时，尼克就去了实验室。
We found somebody who knew about lab stuff. I don't do that. I take care of patients, but I don't do pipettes.
我们找到会做实验的人。我不做实验的。我会看病人，但我不会做分量管。
They taught us how to feed mice drugs. We took mice and we gave them one P, paroxetine.
他们教我们如何喂老鼠吃药。我们给第一组老鼠喂食帕罗西汀，
We gave some other mice pravastatin. And we gave a third group of mice both of them.
给第二组老鼠喂食普伐他汀。第三组的老鼠两种药都喂食。
And lo and behold, glucose went up 20 to 60 milligrams per deciliter in the mice.
惊奇的是，葡萄糖每公升上升20到60毫克，老鼠也有相同的反应。
So the paper was accepted based on the informatics evidence alone,
所以，只有数据证据的报告被接受了，
but we added a little note at the end, saying, oh by the way, if you give these to mice, it goes up.
但我们在最后加了注记说，如果把药物给老鼠，葡萄糖也会上升。
That was great, and the story could have ended there. But I still have six and a half minutes.
太棒了，故事其实就到这里结束。但，我还有六分半钟。
So we were sitting around thinking about all of this, and I don't remember who thought of it, but somebody said,
所以，我们坐下来想一下所有的事，我忘记谁曾经说过，但有人说：
"I wonder if patients who are taking these two drugs are noticing side effects of hyperglycemia.
“不晓得同时服用这两种药的病人，是否有注意到高血糖症的副作用。
They could and they should. How would we ever determine that?"
他们可能知道，也必须知道。我们要如何确定？”
We said, well, what do you do? You're taking a medication, one new medication or two, and you get a funny feeling.
我们说，好吧，你会怎么做？你服用了一种药，一个或两个新药，然后你感觉怪怪的。
What do you do? You go to Google and type in the two drugs you're taking or the one drug you're taking,
你会怎么做？你会去问Google，然后搜寻你在服用的一或两个药名，
and you type in "side effects." What are you experiencing?
然后加上“副作用”。你会找到什么？
So we said OK, let's ask Google if they will share their search logs with us,
所以，我们说，好，我们来问Google能否跟我们分享搜寻纪录，
so that we can look at the search logs and see if patients are doing these kinds of searches.
让我们可以观察搜寻纪录，看是否有病人也在做同样的搜寻。
Google, I am sorry to say, denied our request. So I was bummed.
很抱歉，我得这么说，Google拒绝了我们的请求。所以，我很烦恼。
I was at a dinner with a colleague who works at Microsoft Research and I said,
我跟一个在微软研究室的同事吃晚餐时，我跟他说：
"We wanted to do this study, Google said no, it's kind of a bummer." He said, "Well, we have the Bing searches."
“我们想做这个研究，Google说不行，我有点烦恼。”他说：“我们有Bing搜索引擎啊。”
Yeah. That's great. Now I felt like I was...
是啊！太棒了。现在，我感觉...
I felt like I was talking to Nick again.
我好像又在鼓励尼克一样。
He works for one of the largest companies in the world, and I'm already trying to make him feel better.
他在全世界数一数二的公司上班，我已经开始要安慰他了。
But he said, "No, Russ -- you might not understand.
但他说：“不，洛斯，你可能没搞懂。
We not only have Bing searches, but if you use Internet Explorer to do searches at Google, Yahoo, Bing, any...
我们不只有Bing啊，如果你用IE在Google、雅虎、Bing上搜索...
Then, for 18 months, we keep that data for research purposes only."
之后18个月，我们保留这些数据仅做研究目的使用。
I said, "Now you're talking!" This was Eric Horvitz, my friend at Microsoft.
我说：“这才象话嘛！”这就是我在微软的朋友艾瑞克·霍维兹。
So we did a study where we defined 50 words that a regular person might type in if they're having hyperglycemia,
我们做了一项研究，我们定义出了50个如果一般人有高血糖症时会键入的关键词，
like "fatigue," "loss of appetite," "urinating a lot," "peeing a lot"
像是疲劳、没食欲、频尿等。
forgive me, but that's one of the things you might type in.
请原谅我，但这些就是你可能会键入的关键词。
So we had 50 phrases that we called the "diabetes words." And we did first a baseline.
所以，我们有了50个短语，我们称之为“糖尿病关键词”。我们先设定了一条基线。
And it turns out that about .5 to one percent of all searches on the Internet involve one of those words.
原来，网络上有包含这些关键词的搜寻占了0.5~1%的比例。
So that's our baseline rate. If people type in "paroxetine" or "Paxil" -- those are synonyms
所以，这就是我们的基线率，如果大家键入“帕罗西汀”或“克忧果”--这些是同义字
and one of those words, the rate goes up to about two percent of diabetes-type words,
以及刚刚其中一个关键词，那糖尿病类型的基线率会上升到2%，
if you already know that there's that "paroxetine" word.
如果你已经知道“帕罗西汀”这个字的话。
If it's "pravastatin," the rate goes up to about three percent from the baseline.
如果是“普伐他汀”，那比率会从基线率上升到3％。
If both "paroxetine" and "pravastatin" are present in the query, it goes up to 10 percent,
如果“帕罗西汀”和“普伐他汀”同时出现，那会上升到10%，
a huge three- to four-fold increase in those searches with the two drugs that we were interested in,
有3到4倍的增加，用这两种药搜寻，会出现我们感兴趣的字在里面，
and diabetes-type words or hyperglycemia-type words.
像是糖尿病类的字或高血糖症类的字。
We published this, and it got some attention.
我们发布了这个研究，并得到一些关注。
The reason it deserves attention is that patients are telling us their side effects indirectly through their searches.
它值得被关注的原因是，病人会透过搜寻，直接告诉我们药物的副作用。
We brought this to the attention of the FDA. They were interested.
我们得到了食品药物管理局的关注。他们很感兴趣。
They have set up social media surveillance programs to collaborate with Microsoft,
他们已经成立社会媒体监测计划，与微软展开合作，
which had a nice infrastructure for doing this, and others, to look at Twitter feeds, to look at Facebook feeds,
他们有良好的设备来做这些事，可以观察推特的动态、观察脸书的动态、
to look at search logs, to try to see early signs that drugs, either individually or together, are causing problems.
观察搜寻日志、尝试观察引发问题的无论单一药物或混合药物的早期症状。
What do I take from this? Why tell this story?
我从这件事学到什么？为什么要讲这个故事？
Well, first of all, we have now the promise of big data and medium-sized data
首先，我们现在有大数据及中型数据撑腰，
to help us understand drug interactions and really, fundamentally, drug actions. How do drugs work?
来帮助我们了解药物的相互作用，以及真实、基本的药物作用。药物是如何作用的？
This will create and has created a new ecosystem for understanding how drugs work and to optimize their use.
这将会创造一个新的生态系统，来帮助我们了解药物如何运作以及有效使用它们。
Nick went on; he's a professor at Columbia now. He did this in his PhD for hundreds of pairs of drugs.
尼克继续向前，他现在是哥伦比亚的教授。他用好几百对药物作为博士研究。
He found several very important interactions,
他找到了一些非常重要的药物交互作用，
and so we replicated this and we showed that this is a way that really works for finding drug-drug interactions.
所以，我们复制这个模式，展示出利用这样做来寻找药与药之间的作用真的有效。
However, there's a couple of things. We don't just use pairs of drugs at a time.
然而，还有一些事。我们不会同时一次只服用两种药。
As I said before, there are patients on three, five, seven, nine drugs.
就如我之前所说的，有病人一次是服用三、五、七、九种药。
Have they been studied with respect to their nine-way interaction?
他们有认真研究这九种药的相互作用吗？
Yes, we can do pair-wise, A and B, A and C, A and D,
没错，我们可以做成对的药，A+B、A+C、A+D，
but what about A, B, C, D, E, F, G all together, being taken by the same patient,
但如果同一个病人同时服用ABCDEFG，
perhaps interacting with each other in ways that either makes them more effective or less effective or causes side effects that are unexpected?
那可能会互相产生那些作用？药效更好或更不好？或造成那些意想不到的副作用呢？
We really have no idea. It's a blue sky, open field for us to use data to try to understand the interaction of drugs.
我们真的不知道。这是个开放式的蓝天领域，让我们可以使用数据来尝试了解药物彼此间的作用。
Two more lessons: I want you to think about the power that we were able to generate with the data
另外两件事：我想要各位去想想我们所创造出来的力量，就是我们已经可以
from people who had volunteered their adverse reactions through their pharmacists, through themselves, through their doctors,
透过药剂师、病人本身、病人的医师，来取得自愿者身上他们的药物不良反应，
the people who allowed the databases at Stanford, Harvard, Vanderbilt, to be used for research.
这些人同意他们的数据可以被史丹佛、哈佛、范德堡医学院来做研究使用。
People are worried about data. They're worried about their privacy and security -- they should be.
大家都担心数据问题。他们担心自己的隐私及安全--他们必须要担心。
We need secure systems. But we can't have a system that closes that data off,
我们需要保全系统。但我们不能有一个把数据关起来的系统，
because it is too rich of a source of inspiration, innovation and discovery for new things in medicine.
因为它的资源太丰富了，它对医学界的鼓舞、创新、发现新事物实在太重要了。
And the final thing I want to say is, in this case we found two drugs and it was a little bit of a sad story.
最后，我想说的是，我们发现这两个药的案例，的确是令人难过的故事。
The two drugs actually caused problems. They increased glucose.
这两种药一起服用真的会有问题。同时服用会增加葡萄糖。
They could throw somebody into diabetes who would otherwise not be in diabetes,
会造成一个原本没糖尿病的人发生糖尿病情形。
and so you would want to use the two drugs very carefully together, perhaps not together,
所以，各位如果想一起使用这两种药，一定要非常小心，最好不要一起服用，
make different choices when you're prescribing. But there was another possibility.
当你要开处方签时，看看有没有不同的选择。但也有其他的可能。
We could have found two drugs or three drugs that were interacting in a beneficial way.
我们或许能找到两或三种药，一起服用时也许可以更有效。
We could have found new effects of drugs that neither of them has alone, but together, instead of causing a side effect,
我们或许也可以找到药物本身没有的作用，但在一起服用时不但没有产生副作用，
they could be a new and novel treatment for diseases that don't have treatments or where the treatments are not effective.
反而产生新作用，有可能变成最新的绝症疾病治疗方式，或者原本的治疗方式完全是无效的。
If we think about drug treatment today, all the major breakthroughs
如果我们想想现今的药物治疗方式，所有的重大突破--
for HIV, for tuberculosis, for depression, for diabetes -- it's always a cocktail of drugs.
艾滋病、肺结核、忧郁症，糖尿病--总像是药物鸡尾酒。
And so the upside here, and the subject for a different TED Talk on a different day,
这件事的好处是，也许哪一天，不同的TED主题，
is how can we use the same data sources to find good effects of drugs in combination that will provide us new treatments,
我们又会来到这里分享，我们要如何用同样的数据源来找到药物混用时产生的好效果，它将提供我们新的治疗方式，
new insights into how drugs work and enable us to take care of our patients even better? Thank you very much.
以及对药物如何作用提供新的见解，并且让我们的病人得到更好的照顾。非常感谢各位。