English-Video.net comment policy

The comment field is common to all languages

Let's write in your language and use "Google Translate" together

Please refer to informative community guidelines on TED.com

TEDxNijmegen

Peter van Manen: Better baby care -- thanks to Formula 1

彼得‧范‧梅南: 一級方程式賽車如何幫助寶寶的病情?

Filmed
Views 794,639

一部賽車在一級方程式賽車競賽時,會送出數億個資料數據到修車廠提供即時分析與訊息反饋,那麼何不將如此精細且精密的資料系統運用在其他領域,像是兒童醫院?彼得‧范‧梅南將和我們進一步分享更多資訊。

- Electronic systems expert
Peter van Manen is the Managing Director of McLaren Electronics, which provides data systems to major motorsports series. Full bio

Motor發動機 racing賽跑 is a funny滑稽 old business商業.
賽車是個有趣的老行業
00:12
We make a new car汽車 every一切 year,
每年我們都會造一部新車
00:14
and then we spend the rest休息 of the season季節
而當季的其他時間
00:16
trying to understand理解 what it is we've我們已經 built內置
就用來了解我們達成了那些
00:19
to make it better, to make it faster更快.
然後想辦法突破、讓車子能跑得更快
00:21
And then the next下一個 year, we start開始 again.
下一年,重新開始
00:25
Now, the car汽車 you see in front面前 of you is quite相當 complicated複雜.
大家眼前的這部車結構相當複雜
00:28
The chassis機殼 is made製作 up of about 11,000 components組件,
車子底盤大約由一萬一千組零件所構成
00:32
the engine發動機 another另一個 6,000,
引擎則使用了六千組零件
00:36
the electronics電子產品 about eight and a half thousand.
電子設備將近八千五百組零件
00:38
So there's about 25,000 things there that can go wrong錯誤.
所以總共約有兩萬五千組零件可能出錯
00:41
So motor發動機 racing賽跑 is very much about attention注意 to detail詳情.
因此賽車是非常重視細節的
00:46
The other thing about Formula 1 in particular特定
尤其是一級方程式賽車
00:51
is we're always changing改變 the car汽車.
我們不斷地改車
00:54
We're always trying to make it faster更快.
總想盡辦法讓它能跑得更快
00:56
So every一切 two weeks, we will be making製造
所以每兩週,我們就會製造
00:58
about 5,000 new components組件 to fit適合 to the car汽車.
大約五千組新零件來組裝
01:01
Five to 10 percent百分 of the race種族 car汽車
該年度會有 5% 到 10% 的賽車
01:05
will be different不同 every一切 two weeks of the year.
每兩週就會改頭換面一次
01:08
So how do we do that?
我們是如何辦到的呢?
01:11
Well, we start開始 our life with the racing賽跑 car汽車.
嗯,我們的生活是從賽車開始的
01:14
We have a lot of sensors傳感器 on the car汽車 to measure測量 things.
我們在車子裡安裝了許多感應器來監測車況
01:17
On the race種族 car汽車 in front面前 of you here
你們眼前的這部賽車
01:21
there are about 120 sensors傳感器 when it goes into a race種族.
比賽時裝了約一百廿個感應器
01:23
It's measuring測量 all sorts排序 of things around the car汽車.
來測量整部車的車況
01:26
That data數據 is logged記錄. We're logging記錄 about
資料會被記錄下來,我們在資料系統裡
01:30
500 different不同 parameters參數 within the data數據 systems系統,
記錄了大約五百種不同的參數
01:32
about 13,000 health健康 parameters參數 and events事件
大約一萬三千個健康參數和事件
01:36
to say when things are not working加工 the way they should do,
車子出狀況時,這些紀錄能派上用場
01:39
and we're sending發出 that data數據 back to the garage車庫
我們利用遙測技術
01:44
using運用 telemetry遙測 at a rate of two to four megabits兆位 per second第二.
以每秒 2-4 兆位元的速率
將資料傳回修車廠
01:47
So during a two-hour兩小時 race種族, each car汽車 will be sending發出
所以一場兩小時的比賽中
每部車會傳送出
01:52
750 million百萬 numbers數字.
七億五千萬個數字
01:55
That's twice兩次 as many許多 numbers數字 as words that each of us
這是我們每個人一輩子
01:57
speaks說話 in a lifetime一生.
說話用字的兩倍之多
02:00
It's a huge巨大 amount of data數據.
這是很龐大的資料
02:02
But it's not enough足夠 just to have data數據 and measure測量 it.
但是光有資料或能測量資料是不夠的
02:05
You need to be able能夠 to do something with it.
還必須要懂得如何運用這些資料
02:07
So we've我們已經 spent花費 a lot of time and effort功夫
所以我們花了很多時間和精力
02:09
in turning車削 the data數據 into stories故事
把數據轉換成故事
02:12
to be able能夠 to tell, what's the state of the engine發動機,
可以解讀引擎的狀況如何、
02:14
how are the tires輪胎 degrading降解,
輪胎的耗損情形、
02:17
what's the situation情況 with fuel汽油 consumption消費?
車子油耗又是如何?
02:19
So all of this is taking服用 data數據
所以這些就是紀錄資料
02:23
and turning車削 it into knowledge知識 that we can act法案 upon.
並將其轉換為
我們能採取行動的知識
02:26
Okay, so let's have a look at a little bit of data數據.
好,讓我們來看看一點點資料
02:29
Let's pick a bit of data數據 from
這些資料取自於
02:32
another另一個 three-month-old三個月大的 patient患者.
一個三個月大的病童
02:34
This is a child兒童, and what you're seeing眼看 here is real真實 data數據,
他是個小孩子,大家現在看到的是真實數據
02:37
and on the far right-hand右手 side,
在最右手邊
02:41
where everything starts啟動 getting得到 a little bit catastrophic災難性的,
可以看到情況開始變的有點糟糕
02:43
that is the patient患者 going into cardiac心臟的 arrest逮捕.
這病患的心搏漸漸停止
02:46
It was deemed認為 to be an unpredictable不可預料的 event事件.
這是出乎預料的事情
02:49
This was a heart attack攻擊 that no one could see coming未來.
沒人料想到會心臟病發
02:53
But when we look at the information信息 there,
但是當我們檢查這些資訊時
02:56
we can see that things are starting開始 to become成為
我們可以發現,大約在心搏停止前五分鐘
02:59
a little fuzzy模糊 about five minutes分鐘 or so before the cardiac心臟的 arrest逮捕.
數據開始變得有點模糊不清
03:01
We can see small changes變化
我們可以看到如心搏速率等
03:05
in things like the heart rate moving移動.
有些微的變化
03:07
These were all undetected未被發現 by normal正常 thresholds閾值
這些原本能應用在數據資料上的
03:10
which哪一個 would be applied應用的 to data數據.
都沒有被正常閾值偵測到
03:12
So the question is, why couldn't不能 we see it?
所以問題在於為什麼我們看不到?
03:15
Was this a predictable可預測 event事件?
這是可以預測的事嗎?
03:18
Can we look more at the patterns模式 in the data數據
我們能不能從資料的模式中看出端倪
03:20
to be able能夠 to do things better?
並採取更好的行動呢?
03:23
So this is a child兒童,
這個小孩
03:27
about the same相同 age年齡 as the racing賽跑 car汽車 on stage階段,
年紀跟台上這輛賽車相仿
03:29
three months個月 old.
都是三個月大
03:33
It's a patient患者 with a heart problem問題.
這是一名心臟疾病患者
03:34
Now, when you look at some of the data數據 on the screen屏幕 above以上,
現在你看看螢幕上方的資料
03:37
things like heart rate, pulse脈衝, oxygen, respiration呼吸 rates利率,
心跳速率、脈搏、含氧量、呼吸速率
03:40
they're all unusual異常 for a normal正常 child兒童,
都不是一個正常小朋友該有的數據
03:45
but they're quite相當 normal正常 for the child兒童 there,
但是對這個小朋友卻相當正常
03:48
and so one of the challenges挑戰 you have in health健康 care關心 is,
因此在醫療照護上所面臨的挑戰之一是
03:51
how can I look at the patient患者 in front面前 of me,
如何判斷眼前的病人
03:55
have something which哪一個 is specific具體 for her,
有她專屬的資料模式
03:58
and be able能夠 to detect檢測 when things start開始 to change更改,
並在情況改變、病情開始惡化時
04:01
when things start開始 to deteriorate惡化?
有所察覺?
04:04
Because like a racing賽跑 car汽車, any patient患者,
因為跟賽車的道理一樣,任何患者
04:06
when things start開始 to go bad, you have a short time
遇到緊急狀況時,我們只有極短的時間
04:09
to make a difference區別.
扭轉命運
04:12
So what we did is we took a data數據 system系統
所以我們使用一個
在一級方程式賽車年度裡
04:14
which哪一個 we run every一切 two weeks of the year in Formula 1
每兩週執行一次的資料系統
04:17
and we installed安裝 it on the hospital醫院 computers電腦
並將其安裝於
04:20
at Birmingham伯明翰 Children's兒童 Hospital醫院.
伯明罕兒童醫院的電腦裡
04:23
We streamed data數據 from the bedside床頭 instruments儀器
我們從小兒科加護病房內的
04:25
in their pediatric小兒科的 intensive集約 care關心
床邊儀器得到數據
04:27
so that we could both look at the data數據 in real真實 time
如此一來我們可以即時掌握資料
04:30
and, more importantly重要的, to store商店 the data數據
而且更重要的是,能夠儲存資料
04:33
so that we could start開始 to learn學習 from it.
讓我們能開始從資料中學習
04:36
And then, we applied應用的 an application應用 on top最佳
然後,我們運用一個可以即時讓我們
04:39
which哪一個 would allow允許 us to tease out the patterns模式 in the data數據
整理出資料模式的應用程式
04:44
in real真實 time so we could see what was happening事件,
我們就能看到發生了什麼事
04:47
so we could determine確定 when things started開始 to change更改.
我們就能判斷什麼時候開始有狀況
04:50
Now, in motor發動機 racing賽跑, we're all a little bit ambitious有雄心,
之於賽車,我們都有點企圖心、
04:54
audacious膽大, a little bit arrogant傲慢 sometimes有時,
有點大膽、有時也有點自負
04:58
so we decided決定 we would also look at the children孩子
所以我們決定當那些小朋友
被送往加護病房的途中
05:00
as they were being存在 transported to intensive集約 care關心.
也要能觀察到他們的資料
05:04
Why should we wait until直到 they arrived到達 in the hospital醫院
為什麼我們要等到他們抵達醫院
05:06
before we started開始 to look?
才開始呢?
05:09
And so we installed安裝 a real-time即時的 link鏈接
因此我們在救護車及醫院之間
05:11
between之間 the ambulance救護車 and the hospital醫院,
安裝了一個即時連結機制
05:14
just using運用 normal正常 3G telephony電話 to send發送 that data數據
只需使用一般 3G 通訊傳送資料
05:16
so that the ambulance救護車 became成為 an extra額外 bed
這樣一來,救護車就等同於
05:20
in intensive集約 care關心.
加護病房裡的床位了
05:23
And then we started開始 looking at the data數據.
然後我們開始看那些數據
05:26
So the wiggly蠕動 lines at the top最佳, all the colors顏色,
上面那些波浪狀的線,所有顏色
05:30
this is the normal正常 sort分類 of data數據 you would see on a monitor監控 --
都是我們在監測器上
會看到的正常數據資料
05:32
heart rate, pulse脈衝, oxygen within the blood血液,
心跳速率、脈搏、血中的含氧量
05:36
and respiration呼吸.
及呼吸速率
05:39
The lines on the bottom底部, the blue藍色 and the red,
底部的藍線和紅線
05:42
these are the interesting有趣 ones那些.
滿有趣的
05:45
The red line is showing展示 an automated自動化 version
紅線代表自動感應的
05:46
of the early warning警告 score得分了
預警分數
05:49
that Birmingham伯明翰 Children's兒童 Hospital醫院 were already已經 running賽跑.
這些資料是伯明罕兒童醫院
已經跑過的數據
05:51
They'd他們會 been running賽跑 that since以來 2008,
他們從 2008 年就開始跑這些數據
05:53
and already已經 have stopped停止 cardiac心臟的 arrests逮捕
而且已經成功阻止了
05:56
and distress苦難 within the hospital醫院.
院內心搏停止和心臟窘迫的案例
05:58
The blue藍色 line is an indication跡象
而藍線呢
06:01
of when patterns模式 start開始 to change更改,
是資料模式開始產生變化的指標
06:03
and immediately立即, before we even started開始
立即地,甚至在我們開始
06:06
putting in clinical臨床 interpretation解釋,
臨床解讀這些數據之前
06:08
we can see that the data數據 is speaking請講 to us.
數據已經在對我們說話
06:10
It's telling告訴 us that something is going wrong錯誤.
告訴我們情況不妙
06:13
The plot情節 with the red and the green綠色 blobs斑點,
這張有紅點與綠點的圖表
06:16
this is plotting繪製 different不同 components組件
標示出相互對立資料的不同成分
06:20
of the data數據 against反對 each other.
標示出相互對立資料的不同成分
06:23
The green綠色 is us learning學習 what is normal正常 for that child兒童.
綠色是我們所知那位病童的正常情況資料
06:25
We call it the cloud of normality常態.
我們稱之為常態雲圖
06:29
And when things start開始 to change更改,
當情況開始轉變、
06:32
when conditions條件 start開始 to deteriorate惡化,
病情開始惡化時
06:34
we move移動 into the red line.
我們會移到紅線上
06:37
There's no rocket火箭 science科學 here.
這完全不涉及艱深的學問
06:39
It is displaying顯示 data數據 that exists存在 already已經 in a different不同 way,
只是以不同的方式呈現既有的資料
06:41
to amplify放大 it, to provide提供 cues線索 to the doctors醫生,
詳述它、提供醫生護士更多線索
06:45
to the nurses護士, so they can see what's happening事件.
他們就能了解正在發生什麼事
06:48
In the same相同 way that a good racing賽跑 driver司機
同樣地,一位優秀的賽車手
06:51
relies依賴 on cues線索 to decide決定 when to apply應用 the brakes剎車,
也仰賴線索來決定他何時該煞車、
06:54
when to turn into a corner,
何時該轉彎
06:58
we need to help our physicians醫師 and our nurses護士
我們須要幫助我們的醫生和護士
06:59
to see when things are starting開始 to go wrong錯誤.
讓他們在情況不妙時能夠察覺
07:02
So we have a very ambitious有雄心 program程序.
所以我們有一個野心勃勃的計畫
07:06
We think that the race種族 is on to do something differently不同.
我們認為賽車該採取不同的行動了
07:09
We are thinking思維 big. It's the right thing to do.
我們目標很高,這是對的事情
07:14
We have an approach途徑 which哪一個, if it's successful成功,
我們有這個方法,而且如果它成功了
07:17
there's no reason原因 why it should stay within a hospital醫院.
我們就沒理由只在醫院裡運用它
07:20
It can go beyond the walls牆壁.
這方法能走出醫院之牆
07:22
With wireless無線 connectivity連接 these days,
現在到處都有無線連結
07:24
there is no reason原因 why patients耐心, doctors醫生 and nurses護士
沒有理由病人、醫生和護士
07:26
always have to be in the same相同 place地點
為何總是必須同時在同一個地方
07:30
at the same相同 time.
為何總是必須同時在同一個地方
07:32
And meanwhile與此同時, we'll take our little three-month-old三個月大的 baby寶寶,
此刻,我們會讓這個三個月大的寶寶
07:34
keep taking服用 it to the track跟踪, keeping保持 it safe安全,
安安全全地待在軌道上
07:38
and making製造 it faster更快 and better.
跑得更快更好
07:42
Thank you very much.
謝謝大家
07:44
(Applause掌聲)
(掌聲)
07:45
Translated by Iris Chung
Reviewed by Jessie Lee

▲Back to top

About the speaker:

Peter van Manen - Electronic systems expert
Peter van Manen is the Managing Director of McLaren Electronics, which provides data systems to major motorsports series.

Why you should listen

To say that Peter van Manen has a high-speed job would be an understatement. As Managing Director of McLaren Electronics, which provides electronics and data collection software to motorsports events, he and his team work in real time during a race to improve cars on about 500 different parameters. That's about 750 million data points in two hours.

But recently van Manen and his team have been wondering: Why can't the extremely precise and subtle data-collection and analysis systems used in motorsports be applied elsewhere, for the benefit of all? They have applied their systems to ICU units at Birmingham Children's Hospital with real-time analysis that allows them to proactively prevent cardiac arrests. The unit has seen a 25 percent decrease in life-threatening events. And it's just the beginning.

More profile about the speaker
Peter van Manen | Speaker | TED.com