0:00:00.336,0:00:02.569
Motor racing is a funny old business.

0:00:02.593,0:00:04.886
We make a new car every year,

0:00:04.910,0:00:07.074
and then we spend the rest of the season

0:00:07.098,0:00:09.850
trying to understand[br]what it is we've built

0:00:09.874,0:00:13.071
to make it better, to make it faster.

0:00:13.095,0:00:16.346
And then the next year, we start again.

0:00:16.370,0:00:20.584
Now, the car you see in front[br]of you is quite complicated.

0:00:20.608,0:00:24.203
The chassis is made[br]up of about 11,000 components,

0:00:24.227,0:00:26.671
the engine another 6,000,

0:00:26.695,0:00:29.764
the electronics[br]about eight and a half thousand.

0:00:29.788,0:00:34.165
So there's about 25,000 things[br]there that can go wrong.

0:00:34.189,0:00:38.991
So motor racing is very much[br]about attention to detail.

0:00:39.015,0:00:42.254
The other thing about Formula[br]1 in particular

0:00:42.278,0:00:44.378
is we're always changing the car.

0:00:44.402,0:00:46.658
We're always trying to make it faster.

0:00:46.682,0:00:49.642
So every two weeks, we will be making

0:00:49.666,0:00:53.842
about 5,000 new components[br]to fit to the car.

0:00:53.866,0:00:56.020
Five to 10 percent of the race car

0:00:56.044,0:00:59.772
will be different[br]every two weeks of the year.

0:00:59.796,0:01:02.081
So how do we do that?

0:01:02.105,0:01:05.825
Well, we start our life[br]with the racing car.

0:01:05.849,0:01:09.816
We have a lot of sensors[br]on the car to measure things.

0:01:09.840,0:01:11.698
On the race car in front of you here

0:01:11.722,0:01:14.857
there are about 120 sensors[br]when it goes into a race.

0:01:14.881,0:01:18.509
It's measuring all sorts[br]of things around the car.

0:01:18.533,0:01:20.561
That data is logged. We're logging about

0:01:20.585,0:01:24.265
500 different parameters[br]within the data systems,

0:01:24.289,0:01:27.930
about 13,000 health parameters and events

0:01:27.954,0:01:32.495
to say when things are not[br]working the way they should do,

0:01:32.519,0:01:35.320
and we're sending that data[br]back to the garage

0:01:35.344,0:01:40.299
using telemetry at a rate[br]of two to four megabits per second.

0:01:40.323,0:01:43.426
So during a two-hour race,[br]each car will be sending

0:01:43.450,0:01:45.701
750 million numbers.

0:01:45.725,0:01:48.844
That's twice as many numbers[br]as words that each of us

0:01:48.868,0:01:50.475
speaks in a lifetime.

0:01:50.499,0:01:53.093
It's a huge amount of data.

0:01:53.117,0:01:55.738
But it's not enough just[br]to have data and measure it.

0:01:55.762,0:01:57.896
You need to be able to do[br]something with it.

0:01:57.920,0:02:00.290
So we've spent a lot of time and effort

0:02:00.314,0:02:02.159
in turning the data into stories

0:02:02.183,0:02:05.264
to be able to tell,[br]what's the state of the engine,

0:02:05.288,0:02:07.536
how are the tires degrading,

0:02:07.560,0:02:11.284
what's the situation[br]with fuel consumption?

0:02:11.308,0:02:14.072
So all of this is taking data

0:02:14.096,0:02:17.874
and turning it into knowledge[br]that we can act upon.

0:02:17.898,0:02:20.512
Okay, so let's have a look[br]at a little bit of data.

0:02:20.536,0:02:22.542
Let's pick a bit of data from

0:02:22.566,0:02:25.621
another three-month-old patient.

0:02:25.645,0:02:29.792
This is a child, and what you're[br]seeing here is real data,

0:02:29.816,0:02:31.769
and on the far right-hand side,

0:02:31.793,0:02:34.555
where everything starts getting[br]a little bit catastrophic,

0:02:34.579,0:02:37.940
that is the patient going[br]into cardiac arrest.

0:02:37.964,0:02:41.172
It was deemed to be[br]an unpredictable event.

0:02:41.196,0:02:44.961
This was a heart attack[br]that no one could see coming.

0:02:44.985,0:02:47.511
But when we look at the information there,

0:02:47.535,0:02:49.860
we can see that things[br]are starting to become

0:02:49.884,0:02:53.889
a little fuzzy about five minutes[br]or so before the cardiac arrest.

0:02:53.913,0:02:55.926
We can see small changes

0:02:55.950,0:02:58.309
in things like the heart rate moving.

0:02:58.333,0:03:00.795
These were all undetected[br]by normal thresholds

0:03:00.819,0:03:03.203
which would be applied to data.

0:03:03.227,0:03:06.346
So the question is, why[br]couldn't we see it?

0:03:06.370,0:03:08.927
Was this a predictable event?

0:03:08.951,0:03:11.937
Can we look more[br]at the patterns in the data

0:03:11.961,0:03:15.317
to be able to do things better?

0:03:15.341,0:03:17.967
So this is a child,

0:03:17.991,0:03:21.199
about the same age[br]as the racing car on stage,

0:03:21.223,0:03:22.829
three months old.

0:03:22.853,0:03:25.434
It's a patient with a heart problem.

0:03:25.458,0:03:28.902
Now, when you look at some of the data[br]on the screen above,

0:03:28.926,0:03:33.804
things like heart rate, pulse,[br]oxygen, respiration rates,

0:03:33.828,0:03:36.880
they're all unusual for a normal child,

0:03:36.904,0:03:39.522
but they're quite normal[br]for the child there,

0:03:39.546,0:03:43.660
and so one of the challenges[br]you have in health care is,

0:03:43.684,0:03:46.511
how can I look at the patient[br]in front of me,

0:03:46.535,0:03:49.558
have something which is specific for her,

0:03:49.582,0:03:52.346
and be able to detect when[br]things start to change,

0:03:52.370,0:03:54.445
when things start to deteriorate?

0:03:54.469,0:03:57.495
Because like a racing car, any patient,

0:03:57.519,0:04:00.471
when things start to go[br]bad, you have a short time

0:04:00.495,0:04:02.302
to make a difference.

0:04:02.326,0:04:05.056
So what we did is we took a data system

0:04:05.080,0:04:08.187
which we run every two weeks[br]of the year in Formula 1

0:04:08.211,0:04:11.189
and we installed it[br]on the hospital computers

0:04:11.213,0:04:13.479
at Birmingham Children's Hospital.

0:04:13.503,0:04:15.918
We streamed data[br]from the bedside instruments

0:04:15.942,0:04:18.475
in their pediatric intensive care

0:04:18.499,0:04:21.930
so that we could both look[br]at the data in real time

0:04:21.954,0:04:24.802
and, more importantly, to store the data

0:04:24.826,0:04:27.859
so that we could start to learn from it.

0:04:27.883,0:04:32.243
And then, we applied an application on top

0:04:32.267,0:04:35.513
which would allow us to tease[br]out the patterns in the data

0:04:35.537,0:04:38.469
in real time so we could[br]see what was happening,

0:04:38.493,0:04:42.182
so we could determine when[br]things started to change.

0:04:42.206,0:04:46.045
Now, in motor racing, we're[br]all a little bit ambitious,

0:04:46.069,0:04:48.594
audacious, a little bit[br]arrogant sometimes,

0:04:48.618,0:04:51.992
so we decided we would also[br]look at the children

0:04:52.016,0:04:54.949
as they were being transported[br]to intensive care.

0:04:54.973,0:04:57.497
Why should we wait[br]until they arrived in the hospital

0:04:57.521,0:04:59.097
before we started to look?

0:04:59.121,0:05:02.094
And so we installed a real-time link

0:05:02.118,0:05:04.930
between the ambulance and the hospital,

0:05:04.954,0:05:08.706
just using normal 3G[br]telephony to send that data

0:05:08.730,0:05:11.193
so that the ambulance became an extra bed

0:05:11.217,0:05:14.329
in intensive care.

0:05:14.353,0:05:18.031
And then we started looking at the data.

0:05:18.055,0:05:20.952
So the wiggly lines[br]at the top, all the colors,

0:05:20.976,0:05:24.146
this is the normal sort of data[br]you would see on a monitor --

0:05:24.170,0:05:27.918
heart rate, pulse,[br]oxygen within the blood,

0:05:27.942,0:05:30.553
and respiration.

0:05:30.577,0:05:33.306
The lines on the bottom,[br]the blue and the red,

0:05:33.330,0:05:34.807
these are the interesting ones.

0:05:34.831,0:05:37.875
The red line is showing[br]an automated version

0:05:37.899,0:05:39.472
of the early warning score

0:05:39.496,0:05:42.211
that Birmingham Children's Hospital[br]were already running.

0:05:42.235,0:05:44.297
They'd been running that since 2008,

0:05:44.321,0:05:46.553
and already have stopped cardiac arrests

0:05:46.577,0:05:49.310
and distress within the hospital.

0:05:49.334,0:05:51.742
The blue line is an indication

0:05:51.766,0:05:54.242
of when patterns start to change,

0:05:54.266,0:05:56.551
and immediately, before we even started

0:05:56.575,0:05:58.259
putting in clinical interpretation,

0:05:58.283,0:06:01.129
we can see that the data[br]is speaking to us.

0:06:01.153,0:06:04.665
It's telling us that something[br]is going wrong.

0:06:04.689,0:06:08.481
The plot with the red and the green blobs,

0:06:08.505,0:06:11.286
this is plotting different components

0:06:11.310,0:06:13.833
of the data against each other.

0:06:13.857,0:06:17.673
The green is us learning[br]what is normal for that child.

0:06:17.697,0:06:20.283
We call it the cloud of normality.

0:06:20.307,0:06:22.524
And when things start to change,

0:06:22.548,0:06:25.088
when conditions start to deteriorate,

0:06:25.112,0:06:27.326
we move into the red line.

0:06:27.350,0:06:28.983
There's no rocket science here.

0:06:29.007,0:06:33.096
It is displaying data that exists[br]already in a different way,

0:06:33.120,0:06:36.487
to amplify it, to provide[br]cues to the doctors,

0:06:36.511,0:06:39.225
to the nurses, so they can[br]see what's happening.

0:06:39.249,0:06:42.355
In the same way that a good racing driver

0:06:42.379,0:06:46.399
relies on cues to decide[br]when to apply the brakes,

0:06:46.423,0:06:47.875
when to turn into a corner,

0:06:47.899,0:06:50.793
we need to help[br]our physicians and our nurses

0:06:50.817,0:06:54.413
to see when things[br]are starting to go wrong.

0:06:54.437,0:06:57.359
So we have a very ambitious program.

0:06:57.383,0:07:02.095
We think that the race is on to do[br]something differently.

0:07:02.119,0:07:04.999
We are thinking big.[br]It's the right thing to do.

0:07:05.023,0:07:08.411
We have an approach which,[br]if it's successful,

0:07:08.435,0:07:11.055
there's no reason why it[br]should stay within a hospital.

0:07:11.079,0:07:12.783
It can go beyond the walls.

0:07:12.807,0:07:14.854
With wireless connectivity these days,

0:07:14.878,0:07:18.298
there is no reason why[br]patients, doctors and nurses

0:07:18.322,0:07:20.469
always have to be in the same place

0:07:20.493,0:07:22.462
at the same time.

0:07:22.486,0:07:26.457
And meanwhile, we'll take[br]our little three-month-old baby,

0:07:26.481,0:07:30.214
keep taking it to the track,[br]keeping it safe,

0:07:30.238,0:07:32.547
and making it faster and better.

0:07:32.571,0:07:33.952
Thank you very much.

0:07:33.976,0:07:38.930
(Applause)