ADM-201 dump PMP dumps pdf SSCP exam materials CBAP exam sample questions

一种检测癫痫发作的人工智能手表 – 译学馆
未登陆,请登陆后再发表信息
最新评论 (0)
播放视频

一种检测癫痫发作的人工智能手表

An AI smartwatch that detects seizures | Rosalind Picard

This is Henry, a cute boy.
这是亨利 一个可爱的男孩
And when Henry was three,
当他三岁时
his mom found him having some febrile seizures.
他妈妈发现他有些热性癫痫的症状
Febrile seizures are seizures that occur when you also have a fever.
热性癫痫是发作时伴随着发烧的癫痫病
And the doctor said: “don’t worry too much,kids usually outgrow these.”
医生说:“不用太担心 小孩长大点这种病就好了”
When he was four, he had a convulsive seizure,
当他四岁时 他患上了痉挛性癫痫
the kind that you lose consciousness and shake,
会失去意识并抖动的那种
a generalized tonic-clonic seizure.
全身强直阵挛发作
And while the diagnosis of epilepsy was in the mail,
当癫痫诊断书还在邮寄途中时
Henry’s mom went to get him out of bed one morning,
某天早上 亨利的妈妈去叫他起床
and as she went in his room,
当她走进他的房间
she found his cold, lifeless body.
她发现他身体冰凉 没有生命迹象
Henry died of SUDEP,
亨利死于SUDEP
sudden unexpected death in epilepsy.
即癫痫猝死症
I’m curious how many of you have heard of SUDEP.
我好奇在座有多少人听过癫痫猝死症
This is a very well-educated audience,
各位都是受过良好教育的观众
and I see only a few hands.
但我只看到几个人举手
SUDEP is when an otherwise healthy person with epilepsy dies
癫痫猝死症指的是健康的癫痫患者死亡时
and they can’t attribute it to anything they can find in an autopsy.
无法在尸检中找到病因
There is a SUDEP every seven to nine minutes.
每七到九分钟就有一个人发生癫痫猝死症
That’s on average two per TED Talk.
平均每场TED演讲就有两人
Now, a normal brain has electrical activity.
正常的大脑会产生脑电波
You can see some of the electrical waves coming out of this picture of a brain here.
你可以从这张大脑图片上看到脑电波
And these should look like typical electrical activity
这是脑电图描记器可以直接读到的
that an EEG could read on the surface.
典型的脑电活动
When you have a seizure,
当癫痫发作时
it’s a bit of unusual electrical activity.
就会有异常的脑电活动
And it can be focal,
这可能就是病灶所在
It can take place in just a small part of your brain.
它可以只发生在大脑的一小块区域
When that happens,
当它发生时
you might have a strange sensation.
你可能会有奇怪的感觉
One could be happening several could be happening here in the audience right now,
在座的观众当中可能有一些人现在就有这个状况
and the person next to you might not even know.
而你旁边的人可能完全不知道
However, if you have a seizure where that little brush fire
然而 如果你脑中癫痫发作的病灶从小小的火苗
spreads like a forest fire over the brain,
扩散到像森林大火那样
then it generalizes.
癫痫就会变成全身性的
And that generalized seizure
全身性的发作
takes your consciousness away
会令你失去意识
and cause you to convulse.
并让你抽搐
There are more SUDEPs in the United States every year
在美国 每年因癫痫猝死的人数
than sudden infant death syndrome.
高于因婴儿猝死综合症(死亡的人数)
Now, how many of you have heard of sudden infant death syndrome?
在座有多少人听过婴儿猝死综合症?
All right,pretty much every hand goes up.
好的 几乎人人都举手了
So, what’s going on here?
那么 这是怎么回事?
Why is this so much more common
为什么这种疾病这么普遍
and yet people haven’t heard of it?
然而大家却没听说过它呢?
And what can you do to prevent it?
你要如何预防它呢?
All the two things scientifically shown,
科学指出有两个方法
that prevent or reduce the risk of SUDEP.
可以预防或降低癫痫猝死症的风险
The first is, follow your doctor’s instructions, take your medications.”
第一 谨遵医嘱 按时吃药
Two-thirds of people who have epilepsy
三分之二的癫痫症患者
get it under control with their medications.
都能用药物来控制
The second thing that reduces the risk of SUDEP is companionship.
能降低癫痫猝死症风险的第二个方法就是 陪伴
It’s having somebody there at the time that you have a seizure.
也就是当你发作的时候 要有人在身边
Now, SUDEP, even through most of you have never heard of it,
虽然大部分人没听过癫痫猝死症
is actually the number two cause of years of potential life lost of all neurological disorders.
但它其实是所有神经疾病中最有可能致死的第二名
The vertical axis is the number of deaths
纵轴是死亡人数
times the remaining life span,
乘上剩下的生命时长
so, higher is much worse impact.
那么 越高表示影响越糟糕
SUDEP, however, unlike these others,
然而 癫痫猝死症不像其他疾病
is something that people right here
人们能做点什么
could do something to push that down.
来压制住它
Now, what is a Roz Picard,
为什么罗兹•皮卡德
AI researcher, doing here telling you about SUDEP, right? I’m not a neurologist.
人工智能研究者 要在这里跟大家谈癫痫猝死症?我不是神经病学家
When I was working at the Media Lab
我在麻省理工媒体实验室
on measurement of emotion,
研究情绪测量时
trying to make our machines more intelligent about our emotions,
试图让我们的机器更好的判断识别我们的情绪
we started doing a lot of work measuring stress.
我们开始努力去测量压力
We build lots of sensors
我设置了许多感应器
that measured it in lots of different ways.
用许多不同的方式来测量压力
But one of them in particular,
但其中有种方式很特别
grew out of some of this very old work
是用电信号这种古老的方式
with measuring sweaty palms with an electrical signal.
来测量掌心流汗
This is a signal of skin conductance
这是皮肤电导的信号
that’s known to go up when you get nervous,
大家都知道 当你紧张时 这种信号会上升
but it turns out also goes up with
但结果发现
a lot of other interesting conditions.
还有许多其他有趣的情况也会让它上升
But measuring it with wires on our hand is really inconvenient.
但手上连接电线来测量实在不太方便
So we invented a bunch of other ways of doing this at the MIT Media Lab.
所以 我们在麻省理工媒体实验室里研究出了许多其他方法来做这件事
And with these wearables,
有了这些可穿戴的装备
we started to collect the first-ever clinical quality data 24-7.
我们开始能取得一周七天 一天二十四小时的高质量临床资料 这是前所未有的
Here’s a picture of what that looked like
这张图上的是
the first time an MIT student collected
第一次从麻省理工学院的学生那里收集的
skin conductance on the wrist 24-7.
一周七天 一天二十四小时的手腕上的皮肤电导资料
Let’s zoom in a little bit here.
咱们把它放大一点
What you see is 24 hours from left to right,
从左到右是二十四小时的情况
and here is two days of data.
图上的是两天的资料
And first, what surprised us was sleep
首先 让我们惊讶的是
was the biggest peak of the day.
睡觉时是一天中峰值发生的时间
Now, that’s sounds broken, right?
听起来是出故障了 对吧?
How? I mean you’re calm when you’re asleep,
怎么会呢 我是说 你睡着时是很平静的
so what’s going on here?
那这是怎么回事?
Well, it turns out that our physiology during sleep
结果发现 睡眠期间的生理机能
is very diffrent than our physiology during awake,
和清醒期间非常不同
and while there’s still a bit of a mystery
虽然对于为什么一天之中的峰值
why these peaks are usually the biggest of the day during sleep.
通常出现在睡眠期间 仍然有点难以理解
we now believe they’re related to
但我们现在相信
memory consolidation
它们和记忆的巩固
and memory formation during sleep.
及睡眠期间记忆的形成有关
We also saw things that were exactly what we expected.
还有一些发现 完全在我们的预料之中
When an MIT student is working hard in the lab or on homeworks,
当麻省理工学院的学生在实验室努力或在做功课时
there is not only emotional stress,
不仅有情绪压力
but there’s cognitive load,
还有认知负荷
and it turns out that cognitive load,
结果表明 认知负荷
cognitive effort, mental engagement,
认知努力 心理参与
excitement about learning something
对学习某件事物感到兴奋
those things also make the signal go up.
上述这些也都会造成信号上升
Unfortunately, the embarrassment of we MIT professors,
很遗憾 让我们麻省理工学院教授们尴尬的是
the low point every day is classroom activity.
一天的低点出现在课堂上
Now, I am just showing you one person’s data here,
现在 各位在这里看到的只是一个人的资料
but this, unfortunately is true in general.
但遗憾的是 这也是个普遍现象
This sweatband
这条防汗带
has inside it a homebuilt skin-conductance sensor.
里面有自制的皮肤电导感测器
And one day, one of our undergrads knocked on my door
有一天 我们的一位大学生来敲我的门
right at the end of the December semester,
那时是十二月份学期末
and he said:”Professor Picard, can I please borrow one of your wristband sensors?
他说:“皮卡德教授 能借给我一个腕带传感器吗?
My little brother has autism,
我弟弟有自闭症
he can’t talk, and I want to see what’s streesing him out.”
他不能说话 我想看看是什么使他紧张”
And I said,” Sure, in fact, don’t just take one, take two because they broke easily back then.”
我说:“当然可以 别只拿一条 拿两条 因为它们很容易坏”
So he took them home,
于是 他把腕带传感器带回了家
he put them on his little brother.
并给他弟弟戴上
Now, I was back in MIT, looking at the data on my laptop.
我回到麻省理工大学并在笔记本上看着数据
And the first day,
第一天
I thought, “um, that’s odd, he put them on both wrists
我想着 “哦 很奇怪 他把传感器同时放在了两个胳膊腕上
instead of waiting for one to break.
而不是等一条坏了再换上另一条
Ok, fine, don’t follow my instructions.”
好吧 别听我的指示”
I’m glad he didn’t.
我很开心他没听我的
Second day–chill. Looked like classroom activity.
第二天——扫兴 看起来和课堂上一样
A few more days ahead.
又过了几天
The next day,
接下来的一天
one wrist signal was flat
一条腕上的信号平稳
and the other had the biggest peak I’ve ever seen,
另一条腕上出现了我见过的峰顶
and I thought, “what’s going on?
我在想“发生了什么?
We’ve stressed people out at MIT every way imaginable.
我们在MIT采用了所有可以想到的办法给人们施压
I’ve never seen a peak this big.”
也从未见过这么高的峰顶”
And it was only on one side.
而且只有一边有这种变化
How can you be stressed on one side of your body and not the other?
怎么会只有一边有压力而另一边没事呢?
So I thought one or both sensors must be broken.
所以我认为 其中一个或这两个感应器都出了问题
Now, I’m an electro engineer by training,
现在经过训练 我成了电镀工程师
so I started a whole bunch of stuff to try to debug this,
于是 我开始采用一切措施去调试感应器
and long story short,
长话短说
I could not reproduce this.
我没能使之重现
So I resorted to old-fashioned debugging.
所以我采取老办法进行调试
I called the student at home on vacation.
假期的时候 我在家给学生打了电话
“Hi, how’s your little brother?
“嗨 你弟弟怎么样了?
How’s your Christmas?
圣诞节过得怎么样?
Hey, do you have any idea what happened to him?”
嘿 你对他身上发生的情况有什么看法?”
And I give this particular date and time in data.
我在数据中给出了这个特定的日期和时间
And he said,
他说
“I don’t know, I’ll check the diary.”
“我不知道 我要查一下日记”
Diary? An MIT student keeps a diary?
日记?一个MIT学生还坚持写日记?
So I waited and he came back had exact data and time,
所以我等着 他把准确的数据和时间记录拿来了
and he says, “that was right before he had a grand mal seizure.”
他说“没错 在他发病之前 一切都很正常”
Now, at the time, I didn’t know anything about epilepsy,
那个时候 我对癫痫并不了解
and did a bunch of research,
于是做了大量的研究
realized that another student’s dad is chief of neurosurgery at Children’s Hospital Boston,
想到另一个学生的父亲是波士顿儿童医院的神经外科的主治医生
screwed up my courage and called Dr. Joe Madsen.
我鼓足勇气打了电话给马德森医生
“Hi, Dr. Madsen, my name’s Rosalind Picard.
“嗨 马德森医生 我叫皮卡德
Is it possible somebody could have
有没有可能某些人
a huge sympathetic nervous system surge
在突然发病前
that’s what drives the skin conductance
交感神经系统出现20分钟的巨大波动
20 minutes before a seizure?”
促使皮肤电导系数突然高涨”
And he says:”probably not.”
他说:“可能不会”
He says, “It’s interesting,we’ve had people whose hair stands on end on one arm
他说“这很有趣 我们有个人在发病前
20 minutes before a seizure.”
一只胳膊上的汗毛突然竖了20分钟
And I’m like, “On one arm?”
我在想“一只胳膊?”
I did’t want to tell him that, initially,
起初 我不想告诉他
because I thought this was too ridiculous.
因为我认为这太不可思议了
He explained how this could happen in the brain,
他向我解释了大脑是如何产生这种反应的
and he got interested.
他对此很感兴趣
I showed him the data we made a whole bunch more devices,
我把数据向他展示了 我们制作了大量的装置
got them safety certified.
确保安全
90 families were being enrolled in a study,
90个家庭参加了这项研究
all with children who were going to be monitored 24-7
包括他们的孩子 七天每天24小时
with gold-standard EEG on their scalp
用最标准的EEG放在他们的头上
for reading the brain activity,
对他们的大脑活动进行监控
video to watch the behavior,
通过视频观看大脑活动
electrocardiogram ECG and now EDA,
心电图还有EDA
electrodermal activity,
皮肤电活动
to see if there was something in this periphery
来观察周围是否有一些
that we could easily pick up, related to a seizure.
我们可以很容易看出跟癫痫发作相关的东西
We found, in 100% of the first batch of grand mal seizures,
我们发现 在第一批癫痫大发作中
this whopper of responses in the skin conductance.
所有的人都出现了这种巨大的皮肤电导反应
The blue in the middle, the boy’s sleep,
中间蓝色的部分 这个男孩睡觉时
is usually the biggest peak of the day.
通常处于一天中的最高峰
These three seizures you see here
你看这儿三次癫痫发作
are popping out of the forest like reswood trees.
就像森林中长出的红色的树
Futhermore, when you couple the skin conductance at the top
进一步来讲 当你将顶部皮肤电导和
with the movement from the wrist,
腕部的运动连在一起的时候
and you get lots of data
你会得到很多数据
and train machine learning and AI on it,
如果把训练机器和AI放在上面
you can build an automated AI that detects these patterns,
就能建立一个全自动的监测这些形式的人工智能
much better than just a shake detector can do.
这样比震动探测器监测的更好
So we realized that we needed to get this out,
我们意识到要解决这个问题
and with PHD work of Ming-ZherPoh
通过Ming-ZherPoh的博士研究
and later great improvements by Empatica,
以及后来Empatica的巨大改进
this has made progress and the seizure drtection is much more accurate.
这使我们监测癫痫的数据精确多了
But we also learned some other things about SUDEP during this.
在这期间 我们也学到了一些关于SUDEP的其他的东西
One thing we learned is that SUDEP is
一个是 我们知道SUDEP
while it’s rare after a generalized tonic-clonic seizure,
在一般强直性阵挛发作后很少发生
that’s when it’s most likely to happen after that type.
但这种类型发作后 最有可能发生
And when it happens, it doesn’t happen during the seizure,
当其发作时 不会出现这种情况
and it doesn’t usually happen immediately afterwards,
突然发病后通常不会立刻发生SUDEP
but imediately afterwards,
但一会儿之后
when the person just seeems very still and quiet,
当这个人静止并安静下来之后
they may go into another phase,
他们可们进入另一个阶段
where the breathing stops,
呼吸停止
and then after the breathing stops,
呼吸停止之后
later the heart stops.
然后 心脏停止跳动
So there’s some time to get somebody there.
所以人们需要在那儿待一段时间
We also learned that there is a region
我们也了解到大脑深处有一个区域
deep in the brain called the amygdale,
叫做杏仁孔
which we had been studying in our emotion research a lot.
我们情绪实验一直在研究这个区域
We have two amygdalas.
我们有两个杏仁孔
And if you stimulate the right one,
如果你刺激右边的杏仁孔
you get a big right skin conductance response.
你会看到右边的皮肤电导有很大的反应
Now, you have to sign up right now for a craniotomy to get this done,
你必须立刻签字同意做开颅术进行治疗
not exactly something we’re going to volunteer to do,
我们不确定要求志愿者具体做什么
but it cause a big right skin conductance response.
但这导致右面皮肤电导产生强烈的反应
Stimulate the left one, big left skin conductance response on the palm.
刺激左边 手掌会有很强烈的皮肤电导反应
And furthemore, when somebody stimulates your amygdala
进一步讲 当某人刺激你的杏仁孔
while you’re sitting there and you might just be working,
这时你就坐在那儿 可能一直在工作
you don’t show any signs of distress,
你不会表现出任何痛苦的迹象
but you stop breathing.
但你停止了呼吸
And you don’t start again
你不会再动了
until somebody stimulates you.
直到某人刺激你
“Hey, Roz, are you there?”
“嘿 罗兹 你在吗?”
And you open your mouth to talk.
你张开嘴说话
As you take that breath to speak,
当你想要呼吸说话的时候
you start breathing again.
你开始再次呼吸
So we had started with work on stress,
所以我们顶着压力开始工作
which had enabled us to build lots of sensors
这样使我们建立许多传感器
that were gathering high quality enough data
置于一处 给我们传送高质量的足够的数据
that we could leave the lab and start to get this in the wild;
我们可以离开实验室 开始在野外收集数据
accidentally found a whopper of a response with the seizure,
我们意外地发现 突然发病是个谎言
neurological activation that can cause a much bigger response
神经病学的活动能比传统的压力因素
than traditional stressors;
造成更大的反应
lots of partnership with hospitals
许多医院的合作企业
and an epilepsy monitoring unit,
和癫痫控制科室
especially Children’s Hospital Boston and the Brigham;
尤其是波士顿儿童医院和布里格姆
and machine learning and AI on top of this
机器和还有上面这个人工智能
to take and collect lots more data
能抓取并收集很多数据
in service of trying to understand these events
这些东西帮助我们理解这些事
and if we could prevent SUDEP.
如果我们能预防SUDEP(就更好了)
This is now commercialized by Empatica,
这就是现在被Empatica商业化的
a start-up that I had the privilege to cofound,
一家新兴的 我有幸与人合作的企业
and the team there has done an amazing job improving the technology
这个团队完成了一项惊人的工作
to make a very beautiful sensor
改善了技术 做出了一种非常漂亮的传感器
that not only tells the time and does
不仅可以报时
steps and sleep and all that good stuff,
计步 睡眠 做其他的事情
but this is running real-time AI and machine learning
而且人工智能和机器可以实时
to detect generalized tonic-clonic seizures
监测全身强直性阵挛性癫痫
and send an alert for help
如果我癫痫发作失去意识
if I were t have a seizure and lose consciousness.
它会发射信号进行求助
This just got FDA-approved
这个东西得到了FDA的认证
as the first smartwatch
它是第一个在神经学上
to get approved in neurology.
得到认证的智能手表
Now, the next slide is what made my skin conductance go up.
现在 下一个情景是使我的皮肤电导上升的东西
One mornig, I’m checking my email
一天早晨 我在检查邮件时
and I see a story from a mom
看见一位妈妈给我发来的一个故事
who said she was in the shower,
她说她在淋浴时
and her phone was on the counter by the shower,
她的手机在淋浴旁边的柜台上
and said her daughter might need her help.
她说她的女儿可能需要她的帮助
So she interrrupts her shower and goes running to her daughter’s bedroom,
于是她停止洗澡跑向她女儿的卧室
and she finds her daughter facedown in bed,
她发现她的女儿脸向着床
blue and not breathing.
脸色发青停止了呼吸
She flips her over human stimulation
她将女儿翻过来
and her daughter takes a breath,
她的女儿恢复了呼吸
ane another breath,and her daughter turns pink and is fine.
脸色恢复了红润并逐渐好转
I think I turned white reading this email.
我想我读完这封邮件时 脸色变得苍白
My first reaponse is, “Oh no, it’s not perfect.
我的第一反应是“不 这不对
The Blurtooth could break, the battery could die.
蓝牙会坏 电池会没电
All the things could go wrong. Don’t rely on this.”
一切都会出问题 不要依赖这个”
And she said, “It’s OK. I know no technology is perfect.
她说“没关系 我知道没有完美的技术
None of us can always be there all the time.
我们没有人能一直在那儿看着
But this, this device plus AI
但这个 这个装置加上AI
enabled me to get there in time to save my daughter’s life.”
使我及时赶到救了我女儿”
Now, I’ve been mentioning children,
现在 我一直在强调孩子们
but SUDEP peaks, actually, among people
但是SUDEP高发人群是
in their 20s, 30s and 40s,
20岁30岁40岁的人群
and the next line I’m going to put up
我要提出的下一条
is probably going to make some people uncomfortable,
可能使一些人不舒服
but it’s less uncomfortale than we’ll all be
但大家要知道 如果这条建议扩展到你认识的某个人
if this list is extended to somebody you know.
我们都会舒服很多
Could this happen to somebody you know?
这会发生在你认识的某个人身上吗?
And the reason I bring up this uncomfortable question
我提一个令人不舒服的问题的原因是
is because one in 26 if you
因为每26个人中就有1个
will have epilepsy at some point,
在某种程度上患有癫痫
and from what I’ve been learning,
从我一直的研究来看
people with epilepsy often don’t tell their friends and their neighbors that they have it.
患有癫痫的人经常不会告诉他们的朋友和邻居
So it you’re willing
所以如果你愿意
to let them use an AI or whatever
让他们使用AI或者
to summon you in a moment of possible need,
无论什么时候当你需要它 它都会满足你的需要
if you would let them know that,
如果你可以让他们知道
you could male a diffrence in their life.
你会在他们的生活中起到重大的影响
Why do all this hard work to build AIs?
为什么要努力建立AI呢?
Couple of reasons here: one is Natashe,
这有几个原因:一个是娜塔莎
the girl who lived,
这个活下来的女孩
and her family wanted me to tell you her name.
她的家人想让我告诉大家她的名字
Another is her family
另一个是她的家人
and the wonderful people out there
和外面那些很好的人
who want to be there to support people
他们是想支持
who have conditions that they’ve felt uncomfortable
那些过去一向别人提及自己的状况
in the past mentioning to others.
就感到不舒服的人
And the other reason is all of you,
另一个原因就是在座的各位
because we have the opportunity
因为我们有机会
to shape the future AI.
在未来改进AI技术
We can actually change it,
我们其实可以改变它
because we are rhe ones building it.
因为是我们发明的它
So let’s build AI
所以我们来利用AI
that makes everybody’s lives better.
让每个人的生活变得更美好吧
Thank you.
谢谢大家

发表评论

译制信息
视频概述

本文主要讲了检测癫痫发作的一款人工智能手表

听录译者

甪里

翻译译者

孙卫卫

审核员

审核员 V

视频来源

https://www.youtube.com/watch?v=YkYdxjaqt58

相关推荐