How can Formula 1 racing help ... babies?
-
0:00 - 0:02Motor racing is a funny old business.
-
0:02 - 0:04We make a new car every year,
-
0:04 - 0:07and then we spend the rest of the season
-
0:07 - 0:09trying to understand what it is we've built
-
0:09 - 0:13to make it better, to make it faster.
-
0:13 - 0:16And then the next year, we start again.
-
0:16 - 0:20Now, the car you see in front of you is quite complicated.
-
0:20 - 0:24The chassis is made up of about 11,000 components,
-
0:24 - 0:26the engine another 6,000,
-
0:26 - 0:29the electronics about eight and a half thousand.
-
0:29 - 0:34So there's about 25,000 things there that can go wrong.
-
0:34 - 0:39So motor racing is very much about attention to detail.
-
0:39 - 0:42The other thing about Formula 1 in particular
-
0:42 - 0:44is we're always changing the car.
-
0:44 - 0:46We're always trying to make it faster.
-
0:46 - 0:49So every two weeks, we will be making
-
0:49 - 0:53about 5,000 new components to fit to the car.
-
0:53 - 0:56Five to 10 percent of the race car
-
0:56 - 0:59will be different every two weeks of the year.
-
0:59 - 1:02So how do we do that?
-
1:02 - 1:05Well, we start our life with the racing car.
-
1:05 - 1:09We have a lot of sensors on the car to measure things.
-
1:09 - 1:11On the race car in front of you here
-
1:11 - 1:14there are about 120 sensors when it goes into a race.
-
1:14 - 1:18It's measuring all sorts of things around the car.
-
1:18 - 1:20That data is logged. We're logging about
-
1:20 - 1:24500 different parameters within the data systems,
-
1:24 - 1:27about 13,000 health parameters and events
-
1:27 - 1:32to say when things are not working the way they should do,
-
1:32 - 1:35and we're sending that data back to the garage
-
1:35 - 1:40using telemetry at a rate of two to four megabits per second.
-
1:40 - 1:43So during a two-hour race, each car will be sending
-
1:43 - 1:45750 million numbers.
-
1:45 - 1:48That's twice as many numbers as words that each of us
-
1:48 - 1:50speaks in a lifetime.
-
1:50 - 1:53It's a huge amount of data.
-
1:53 - 1:55But it's not enough just to have data and measure it.
-
1:55 - 1:57You need to be able to do something with it.
-
1:57 - 2:00So we've spent a lot of time and effort
-
2:00 - 2:02in turning the data into stories
-
2:02 - 2:05to be able to tell, what's the state of the engine,
-
2:05 - 2:07how are the tires degrading,
-
2:07 - 2:11what's the situation with fuel consumption?
-
2:11 - 2:14So all of this is taking data
-
2:14 - 2:17and turning it into knowledge that we can act upon.
-
2:17 - 2:20Okay, so let's have a look at a little bit of data.
-
2:20 - 2:22Let's pick a bit of data from
-
2:22 - 2:25another three-month-old patient.
-
2:25 - 2:29This is a child, and what you're seeing here is real data,
-
2:29 - 2:31and on the far right-hand side,
-
2:31 - 2:34where everything starts getting a little bit catastrophic,
-
2:34 - 2:37that is the patient going into cardiac arrest.
-
2:37 - 2:41It was deemed to be an unpredictable event.
-
2:41 - 2:44This was a heart attack that no one could see coming.
-
2:44 - 2:47But when we look at the information there,
-
2:47 - 2:49we can see that things are starting to become
-
2:49 - 2:53a little fuzzy about five minutes or so before the cardiac arrest.
-
2:53 - 2:55We can see small changes
-
2:55 - 2:58in things like the heart rate moving.
-
2:58 - 3:00These were all undetected by normal thresholds
-
3:00 - 3:03which would be applied to data.
-
3:03 - 3:06So the question is, why couldn't we see it?
-
3:06 - 3:08Was this a predictable event?
-
3:08 - 3:11Can we look more at the patterns in the data
-
3:11 - 3:15to be able to do things better?
-
3:15 - 3:17So this is a child,
-
3:17 - 3:21about the same age as the racing car on stage,
-
3:21 - 3:22three months old.
-
3:22 - 3:25It's a patient with a heart problem.
-
3:25 - 3:28Now, when you look at some of the data on the screen above,
-
3:28 - 3:33things like heart rate, pulse, oxygen, respiration rates,
-
3:33 - 3:36they're all unusual for a normal child,
-
3:36 - 3:39but they're quite normal for the child there,
-
3:39 - 3:43and so one of the challenges you have in health care is,
-
3:43 - 3:46how can I look at the patient in front of me,
-
3:46 - 3:49have something which is specific for her,
-
3:49 - 3:52and be able to detect when things start to change,
-
3:52 - 3:54when things start to deteriorate?
-
3:54 - 3:57Because like a racing car, any patient,
-
3:57 - 4:00when things start to go bad, you have a short time
-
4:00 - 4:02to make a difference.
-
4:02 - 4:05So what we did is we took a data system
-
4:05 - 4:08which we run every two weeks of the year in Formula One
-
4:08 - 4:11and we installed it on the hospital computers
-
4:11 - 4:13at Birmingham Children's Hospital.
-
4:13 - 4:15We streamed data from the bedside instruments
-
4:15 - 4:18in their pediatric intensive care
-
4:18 - 4:21so that we could both look at the data in real time
-
4:21 - 4:24and, more importantly, to store the data
-
4:24 - 4:27so that we could start to learn from it.
-
4:27 - 4:32And then, we applied an application on top
-
4:32 - 4:35which would allow us to tease out the patterns in the data
-
4:35 - 4:38in real time so we could see what was happening,
-
4:38 - 4:42so we could determine when things started to change.
-
4:42 - 4:46Now, in motor racing, we're all a little bit ambitious,
-
4:46 - 4:48audacious, a little bit arrogant sometimes,
-
4:48 - 4:52so we decided we would also look at the children
-
4:52 - 4:54as they were being transported to intensive care.
-
4:54 - 4:57Why should we wait until they arrived in the hospital
-
4:57 - 4:59before we started to look?
-
4:59 - 5:02And so we installed a real-time link
-
5:02 - 5:04between the ambulance and the hospital,
-
5:04 - 5:08just using normal 3G telephony to send that data
-
5:08 - 5:11so that the ambulance became an extra bed
-
5:11 - 5:14in intensive care.
-
5:14 - 5:18And then we started looking at the data.
-
5:18 - 5:20So the wiggly lines at the top, all the colors,
-
5:20 - 5:24this is the normal sort of data you would see on a monitor --
-
5:24 - 5:27heart rate, pulse, oxygen within the blood,
-
5:27 - 5:30and respiration.
-
5:30 - 5:33The lines on the bottom, the blue and the red,
-
5:33 - 5:34these are the interesting ones.
-
5:34 - 5:37The red line is showing an automated version
-
5:37 - 5:39of the early warning score
-
5:39 - 5:41that Birmingham Children's Hospital were already running.
-
5:41 - 5:44They'd been running that since 2008,
-
5:44 - 5:46and already have stopped cardiac arrests
-
5:46 - 5:49and distress within the hospital.
-
5:49 - 5:51The blue line is an indication
-
5:51 - 5:54of when patterns start to change,
-
5:54 - 5:56and immediately, before we even started
-
5:56 - 5:58putting in clinical interpretation,
-
5:58 - 6:01we can see that the data is speaking to us.
-
6:01 - 6:04It's telling us that something is going wrong.
-
6:04 - 6:08The plot with the red and the green blobs,
-
6:08 - 6:11this is plotting different components
-
6:11 - 6:13of the data against each other.
-
6:13 - 6:17The green is us learning what is normal for that child.
-
6:17 - 6:20We call it the cloud of normality.
-
6:20 - 6:22And when things start to change,
-
6:22 - 6:25when conditions start to deteriorate,
-
6:25 - 6:27we move into the red line.
-
6:27 - 6:29There's no rocket science here.
-
6:29 - 6:33It is displaying data that exists already in a different way,
-
6:33 - 6:36to amplify it, to provide cues to the doctors,
-
6:36 - 6:39to the nurses, so they can see what's happening.
-
6:39 - 6:42In the same way that a good racing driver
-
6:42 - 6:46relies on cues to decide when to apply the brakes,
-
6:46 - 6:47when to turn into a corner,
-
6:47 - 6:50we need to help our physicians and our nurses
-
6:50 - 6:54to see when things are starting to go wrong.
-
6:54 - 6:57So we have a very ambitious program.
-
6:57 - 7:02We think that the race is on to do something differently.
-
7:02 - 7:05We are thinking big. It's the right thing to do.
-
7:05 - 7:08We have an approach which, if it's successful,
-
7:08 - 7:10there's no reason why it should stay within a hospital.
-
7:10 - 7:12It can go beyond the walls.
-
7:12 - 7:14With wireless connectivity these days,
-
7:14 - 7:18there is no reason why patients, doctors and nurses
-
7:18 - 7:20always have to be in the same place
-
7:20 - 7:22at the same time.
-
7:22 - 7:26And meanwhile, we'll take our little three-month-old baby,
-
7:26 - 7:30keep taking it to the track, keeping it safe,
-
7:30 - 7:32and making it faster and better.
-
7:32 - 7:33Thank you very much.
-
7:33 - 7:38(Applause)
- Title:
- How can Formula 1 racing help ... babies?
- Speaker:
- Peter van Manen
- Description:
-
During a Formula 1 race, a car sends hundreds of millions of data points to its garage for real-time analysis and feedback. So why not use this detailed and rigorous data system elsewhere, like ... at children’s hospitals? Peter van Manen tells us more.
- Video Language:
- English
- Team:
- closed TED
- Project:
- TEDTalks
- Duration:
- 07:56
Krystian Aparta edited English subtitles for Better baby care -- thanks to Formula 1 | ||
Jenny Zurawell edited English subtitles for Better baby care -- thanks to Formula 1 | ||
Thu-Huong Ha approved English subtitles for Better baby care -- thanks to Formula 1 | ||
Thu-Huong Ha edited English subtitles for Better baby care -- thanks to Formula 1 | ||
Thu-Huong Ha edited English subtitles for Better baby care -- thanks to Formula 1 | ||
Thu-Huong Ha edited English subtitles for Better baby care -- thanks to Formula 1 | ||
Morton Bast accepted English subtitles for Better baby care -- thanks to Formula 1 | ||
Morton Bast edited English subtitles for Better baby care -- thanks to Formula 1 |