WEBVTT

99:59:59.999 --> 99:59:59.999
I have a tendency to assume the worst,

99:59:59.999 --> 99:59:59.999
and once in a while,
this habit plays tricks on me.

99:59:59.999 --> 99:59:59.999
For example, if I feel
unexpected pain in my body

99:59:59.999 --> 99:59:59.999
that I have not experienced before
and that I cannot attribute,

99:59:59.999 --> 99:59:59.999
then all of a sudden, my mind
my turn a tense back into heart disease

99:59:59.999 --> 99:59:59.999
or calf muscle pain into ??.

99:59:59.999 --> 99:59:59.999
But so far I haven't been diagnosed
with any deadly or incurable disease.

99:59:59.999 --> 99:59:59.999
Sometimes things just hurt
for no clear reason.

99:59:59.999 --> 99:59:59.999
But not everyone is as lucky as me.

99:59:59.999 --> 99:59:59.999
Every year, more than
50 million people die worldwide.

99:59:59.999 --> 99:59:59.999
Especially in high-income
economies like ours,

99:59:59.999 --> 99:59:59.999
a large fraction of deaths
is caused by slowly progressing diseases:

99:59:59.999 --> 99:59:59.999
heart disease, chronic lung disease,
cancer, Alzheimer's, diabetes,

99:59:59.999 --> 99:59:59.999
just to name a few.

99:59:59.999 --> 99:59:59.999
Now, humanity has made tremendous progress
in diagnosing and treating many of these,

99:59:59.999 --> 99:59:59.999
but we are at stage
where further advancement in health

99:59:59.999 --> 99:59:59.999
cannot be achieved only
by developing new treatments,

99:59:59.999 --> 99:59:59.999
and this becomes evident
when we look at one aspect

99:59:59.999 --> 99:59:59.999
that many of these
diseases have in common.

99:59:59.999 --> 99:59:59.999
The probability for successful treatment

99:59:59.999 --> 99:59:59.999
strongly depends on
when treatment is started,

99:59:59.999 --> 99:59:59.999
but a disease is typically only detected
once symptoms occur,

99:59:59.999 --> 99:59:59.999
and the problem here is that in fact
many disease can remain asymptomatic,

99:59:59.999 --> 99:59:59.999
hence undetected,
for a long period of time.

99:59:59.999 --> 99:59:59.999
Because of this, there is
a persisting need for new ways

99:59:59.999 --> 99:59:59.999
of detecting disease at early stage,
way before any symptoms occur.

99:59:59.999 --> 99:59:59.999
In health care, this is called screening,

99:59:59.999 --> 99:59:59.999
and as defined by
the World Health Organization,

99:59:59.999 --> 99:59:59.999
screening is the
presumptive identification

99:59:59.999 --> 99:59:59.999
of unrecognized disease

99:59:59.999 --> 99:59:59.999
in an apparently healthy person

99:59:59.999 --> 99:59:59.999
by means of tests that can be applied
rapidly and easily.

99:59:59.999 --> 99:59:59.999
That's a long definition,
so let me repeat it:

99:59:59.999 --> 99:59:59.999
identification of unrecognized disease

99:59:59.999 --> 99:59:59.999
in an apparently healthy person

99:59:59.999 --> 99:59:59.999
by means of tests that can be applied
both rapidly and easily.

99:59:59.999 --> 99:59:59.999
And I want to put special emphasis
on the words "rapidly" and "easily,"

99:59:59.999 --> 99:59:59.999
because many of
the existing screening methods

99:59:59.999 --> 99:59:59.999
are exactly the opposite.

99:59:59.999 --> 99:59:59.999
And those of you who have gone colonoscopy

99:59:59.999 --> 99:59:59.999
as part of a screening program
for colorectal cancer

99:59:59.999 --> 99:59:59.999
will know what I mean.

99:59:59.999 --> 99:59:59.999
Obviously, there's a variety
of medical tools available

99:59:59.999 --> 99:59:59.999
to perform screening tests.

99:59:59.999 --> 99:59:59.999
This ranges from imaging techniques

99:59:59.999 --> 99:59:59.999
such as radiography
or magnetic resonance imaging

99:59:59.999 --> 99:59:59.999
to the analysis of blood or tissue.

99:59:59.999 --> 99:59:59.999
We have all had such tests.

99:59:59.999 --> 99:59:59.999
But there's one medium
that for long has been overlooked,

99:59:59.999 --> 99:59:59.999
a medium that is easily accessible,

99:59:59.999 --> 99:59:59.999
basically non-depletable,

99:59:59.999 --> 99:59:59.999
and it holds tremendous promise
for medical analysis,

99:59:59.999 --> 99:59:59.999
and that is our breath.

99:59:59.999 --> 99:59:59.999
Human breath is essentially
composed of five components:

99:59:59.999 --> 99:59:59.999
nitrogen, oxygen,
carbon dioxide, water and argon,

99:59:59.999 --> 99:59:59.999
but besides these five, there are
hundreds of other components

99:59:59.999 --> 99:59:59.999
that are present in very low quantity.

99:59:59.999 --> 99:59:59.999
These are called volatile
organic compounds,

99:59:59.999 --> 99:59:59.999
and we release hundreds,
even thousands of them,

99:59:59.999 --> 99:59:59.999
every time we exhale.

99:59:59.999 --> 99:59:59.999
The analysis of these volatile
organic compounds in our breath

99:59:59.999 --> 99:59:59.999
is called breath analysis.

99:59:59.999 --> 99:59:59.999
In fact, I believe that many of you
have already experienced breath analysis.

99:59:59.999 --> 99:59:59.999
Imagine, you're driving home late at night

99:59:59.999 --> 99:59:59.999
when suddenly there's
a friendly police officer

99:59:59.999 --> 99:59:59.999
who asks you kindly but firmly

99:59:59.999 --> 99:59:59.999
to pull over and blow
into a device like this one.

99:59:59.999 --> 99:59:59.999
This is an alcohol breath tester

99:59:59.999 --> 99:59:59.999
that is used to measure
the ethanol concentration in your breath

99:59:59.999 --> 99:59:59.999
and determine whether driving
in your condition is a clever idea.

99:59:59.999 --> 99:59:59.999
Now I'd say my driving was pretty good,

99:59:59.999 --> 99:59:59.999
but let me check.

99:59:59.999 --> 99:59:59.999
(Beep)

99:59:59.999 --> 99:59:59.999
0.0, so nothing to worry about, all fine.

99:59:59.999 --> 99:59:59.999
Now, imagine a device like this one

99:59:59.999 --> 99:59:59.999
that does not only measure
alcohol levels in your breath,

99:59:59.999 --> 99:59:59.999
but that detects diseases

99:59:59.999 --> 99:59:59.999
like the ones I've shown you
and potentially many more.

99:59:59.999 --> 99:59:59.999
The concept of correlating
the smell of a person's breath

99:59:59.999 --> 99:59:59.999
with certain medical conditions

99:59:59.999 --> 99:59:59.999
in fact dates back to Ancient Greece,

99:59:59.999 --> 99:59:59.999
but only recently, research efforts
on breath analysis have skyrocketed,

99:59:59.999 --> 99:59:59.999
and what once was a dream
is now becoming reality.

99:59:59.999 --> 99:59:59.999
And let me pull up this list again
that I showed you earlier.

99:59:59.999 --> 99:59:59.999
For the majority of diseases listed here,

99:59:59.999 --> 99:59:59.999
there's substantial scientific evidence
suggesting that the disease

99:59:59.999 --> 99:59:59.999
could be detected by breath analysis.

99:59:59.999 --> 99:59:59.999
But how does it work exactly?

99:59:59.999 --> 99:59:59.999
The essential part is a sensor device

99:59:59.999 --> 99:59:59.999
that detects the volatile
organic compounds in our breath.

99:59:59.999 --> 99:59:59.999
Simply put, when exposed
to a breath sample,

99:59:59.999 --> 99:59:59.999
the sensor outputs a complex signature

99:59:59.999 --> 99:59:59.999
that results from the mixture of volatile
organic compounds that we exhale.

99:59:59.999 --> 99:59:59.999
Now, this signature represents
a fingerprint of your metabolism,

99:59:59.999 --> 99:59:59.999
your microbiome,

99:59:59.999 --> 99:59:59.999
and the biochemical processes
that occur in your body.

99:59:59.999 --> 99:59:59.999
If you have a disease,

99:59:59.999 --> 99:59:59.999
your organisms will change,

99:59:59.999 --> 99:59:59.999
and so will the composition
of your exhaled breath,

99:59:59.999 --> 99:59:59.999
and then the only thing that is left to do

99:59:59.999 --> 99:59:59.999
is to correlate a certain signature

99:59:59.999 --> 99:59:59.999
with the presence or absence
of certain medical conditions.

99:59:59.999 --> 99:59:59.999
The technology promises
several undeniable benefits.

99:59:59.999 --> 99:59:59.999
Firstly, the sensor can be miniaturized

99:59:59.999 --> 99:59:59.999
and integrated into small,
handheld devices

99:59:59.999 --> 99:59:59.999
like this alcohol breath tester.

99:59:59.999 --> 99:59:59.999
This would allow the test to be used
in many different settings

99:59:59.999 --> 99:59:59.999
and even at home,

99:59:59.999 --> 99:59:59.999
so that a visit at a doctor's office

99:59:59.999 --> 99:59:59.999
is not needed each time
a test shall be performed.

99:59:59.999 --> 99:59:59.999
Secondly, breath analysis is non-invasive

99:59:59.999 --> 99:59:59.999
and can be as simple as blowing
into an alcohol breath tester.

99:59:59.999 --> 99:59:59.999
Such simplicity and ease of use
would reduce patient burden

99:59:59.999 --> 99:59:59.999
and provide an incentive
for broad adoption of the technology.

99:59:59.999 --> 99:59:59.999
And thirdly, the technology is so flexible

99:59:59.999 --> 99:59:59.999
that the same device could be used

99:59:59.999 --> 99:59:59.999
to detect a broad range
of medical conditions.

99:59:59.999 --> 99:59:59.999
Breath analysis could be used to screen
for multiple diseases at the same time.

99:59:59.999 --> 99:59:59.999
Nowadays, each disease typically requires

99:59:59.999 --> 99:59:59.999
a different medical tool
to perform a screening test,

99:59:59.999 --> 99:59:59.999
but this means you can only find
what you're looking for.

99:59:59.999 --> 99:59:59.999
With all of these features,
breath analysis is predestined

99:59:59.999 --> 99:59:59.999
to deliver what many traditional
screening tests are lacking,

99:59:59.999 --> 99:59:59.999
and most importantly,
all of these features

99:59:59.999 --> 99:59:59.999
should eventually provide us
with a platform for medical analysis

99:59:59.999 --> 99:59:59.999
that can operate at attractively
low cost per test.

99:59:59.999 --> 99:59:59.999
On the contrary, existing medical tools

99:59:59.999 --> 99:59:59.999
often lead to rather high cost per test.

99:59:59.999 --> 99:59:59.999
Then, in order to keep costs down,

99:59:59.999 --> 99:59:59.999
the number of tests
needs to be restricted,

99:59:59.999 --> 99:59:59.999
and this means a) that the tests
can only be performed

99:59:59.999 --> 99:59:59.999
on a narrow part of the population,
for example the high-risk population,

99:59:59.999 --> 99:59:59.999
and b) that the number of tests per person
needs to be kept at a minimum.

99:59:59.999 --> 99:59:59.999
But wouldn't it actually be beneficial

99:59:59.999 --> 99:59:59.999
if the test was performed
on a larger group of people

99:59:59.999 --> 99:59:59.999
and more often and over a longer period
of time for each individual?

99:59:59.999 --> 99:59:59.999
Especially the latter would give access
to something very valuable

99:59:59.999 --> 99:59:59.999
that is called longitudinal data.

99:59:59.999 --> 99:59:59.999
Longitudinal data is a dataset
that tracks the same patient

99:59:59.999 --> 99:59:59.999
over the course of many months or years.

99:59:59.999 --> 99:59:59.999
Nowadays, medical decisions

99:59:59.999 --> 99:59:59.999
are often based on a limited dataset

99:59:59.999 --> 99:59:59.999
where over a glimpse
of a patient's medical history

99:59:59.999 --> 99:59:59.999
is available for decision-making.

99:59:59.999 --> 99:59:59.999
In such a case,

99:59:59.999 --> 99:59:59.999
abnormalities are typically detected

99:59:59.999 --> 99:59:59.999
by comparing a patient's health profile

99:59:59.999 --> 99:59:59.999
to the average health profile
of a reference population.

99:59:59.999 --> 99:59:59.999
Longitudinal data would
open up a new dimension

99:59:59.999 --> 99:59:59.999
and allow abnormalities to be detected

99:59:59.999 --> 99:59:59.999
based on a patient's own medical history.

99:59:59.999 --> 99:59:59.999
This will pave the way
for personalized treatment.

99:59:59.999 --> 99:59:59.999
Sounds pretty great, right?

99:59:59.999 --> 99:59:59.999
Now you will certainly have a question

99:59:59.999 --> 99:59:59.999
that is something like,

99:59:59.999 --> 99:59:59.999
"If the technology is as great as he says,
then why aren't they using it today?"

99:59:59.999 --> 99:59:59.999
And the only answer I can give you is,

99:59:59.999 --> 99:59:59.999
not everything is as easy as it sounds.

99:59:59.999 --> 99:59:59.999
There are technical challenges.

99:59:59.999 --> 99:59:59.999
For example, there's the need
for extremely reliable sensors

99:59:59.999 --> 99:59:59.999
that can detect mixtures
of volatile organic compounds

99:59:59.999 --> 99:59:59.999
with sufficient reproducibility.

99:59:59.999 --> 99:59:59.999
And another technical challenge is,

99:59:59.999 --> 99:59:59.999
how do you sample a person's breath
in a very defined manner

99:59:59.999 --> 99:59:59.999
so that the sampling process itself

99:59:59.999 --> 99:59:59.999
does not alter the result of the analysis?

99:59:59.999 --> 99:59:59.999
And there's the need for data.

99:59:59.999 --> 99:59:59.999
Breath analysis needs
to be validated in clinical trials,

99:59:59.999 --> 99:59:59.999
and enough data needs to be collected

99:59:59.999 --> 99:59:59.999
so that individual conditions

99:59:59.999 --> 99:59:59.999
can be measured against baselines.

99:59:59.999 --> 99:59:59.999
Breath analysis can only succeed

99:59:59.999 --> 99:59:59.999
if a large enough dataset can be generated

99:59:59.999 --> 99:59:59.999
and made available for broad use.

99:59:59.999 --> 99:59:59.999
If breath analysis
holds up to its promises,

99:59:59.999 --> 99:59:59.999
this is a technology
that could truly aid us

99:59:59.999 --> 99:59:59.999
to transform our health care system,

99:59:59.999 --> 99:59:59.999
transform it from a reactive system

99:59:59.999 --> 99:59:59.999
that treatment is triggered
by symptoms of disease

99:59:59.999 --> 99:59:59.999
to a proactive system

99:59:59.999 --> 99:59:59.999
where disease detection,
diagnosis and treatment

99:59:59.999 --> 99:59:59.999
can happen at early stage,

99:59:59.999 --> 99:59:59.999
way before any symptoms occur.

99:59:59.999 --> 99:59:59.999
Now this brings me to my last point,
and it's a fundamental one.

99:59:59.999 --> 99:59:59.999
What exactly is a disease?

99:59:59.999 --> 99:59:59.999
Imagine that breath analysis
can be commercialized as I describe it

99:59:59.999 --> 99:59:59.999
and early detection becomes routine.

99:59:59.999 --> 99:59:59.999
A problem that remains

99:59:59.999 --> 99:59:59.999
is in fact a problem that
any screening activity has to face,

99:59:59.999 --> 99:59:59.999
because for many diseases,

99:59:59.999 --> 99:59:59.999
it is often impossible to predict
with sufficient certainty

99:59:59.999 --> 99:59:59.999
whether the disease
would ever cause any symptoms

99:59:59.999 --> 99:59:59.999
or put a person's life at risk.

99:59:59.999 --> 99:59:59.999
This is called overdiagnosis,

99:59:59.999 --> 99:59:59.999
and it leads to a dilemma.

99:59:59.999 --> 99:59:59.999
If a disease is identified,

99:59:59.999 --> 99:59:59.999
you could decide not to treat it

99:59:59.999 --> 99:59:59.999
because there's a certain probability
that you would never suffer from it.

99:59:59.999 --> 99:59:59.999
But how much would you suffer

99:59:59.999 --> 99:59:59.999
just from knowing that you have
a potentially deadly disease?

99:59:59.999 --> 99:59:59.999
And wouldn't you actually regret that the
disease was detected in the first place?

99:59:59.999 --> 99:59:59.999
Your second option
is to undergo early treatment

99:59:59.999 --> 99:59:59.999
with the hope for curing it,

99:59:59.999 --> 99:59:59.999
but often this would not
come without side effects.

99:59:59.999 --> 99:59:59.999
To be precise,

99:59:59.999 --> 99:59:59.999
the bigger problem is not overdiagnosis,

99:59:59.999 --> 99:59:59.999
it's overtreatment,

99:59:59.999 --> 99:59:59.999
because not every disease
has to be treated immediately

99:59:59.999 --> 99:59:59.999
just because a treatment is available.

99:59:59.999 --> 99:59:59.999
The increasing adoption
of routine screening

99:59:59.999 --> 99:59:59.999
will raise the question:

99:59:59.999 --> 99:59:59.999
what do we call a disease
that can rationalize treatment,

99:59:59.999 --> 99:59:59.999
and what is just an abnormality
that should not be a source of concern?

99:59:59.999 --> 99:59:59.999
My hopes are that routine screening
using breath analysis

99:59:59.999 --> 99:59:59.999
can provide enough data and insight

99:59:59.999 --> 99:59:59.999
so that at some point,
we'll be able to break this dilemma

99:59:59.999 --> 99:59:59.999
and predict with sufficient certainty

99:59:59.999 --> 99:59:59.999
whether and when to treat at early stage.

99:59:59.999 --> 99:59:59.999
Our breath, and the mixture
of volatile organic compounds

99:59:59.999 --> 99:59:59.999
that we exhale,

99:59:59.999 --> 99:59:59.999
hold tremendous amounts of information
on our physiological condition.

99:59:59.999 --> 99:59:59.999
With what we know today,
we have only scratched the surface.

99:59:59.999 --> 99:59:59.999
As we collect more and more data
and breath profiles across the population,

99:59:59.999 --> 99:59:59.999
including all varieties of gender,
age, origin and lifestyle,

99:59:59.999 --> 99:59:59.999
the power of breath analysis
should increase,

99:59:59.999 --> 99:59:59.999
and eventually breath analysis
should provide us with a powerful tool

99:59:59.999 --> 99:59:59.999
not only to proactively detect
specific diseases

99:59:59.999 --> 99:59:59.999
but to predict and
ultimately prevent them,

99:59:59.999 --> 99:59:59.999
and this should be enough
motivation to embrace

99:59:59.999 --> 99:59:59.999
the opportunities and challenges

99:59:59.999 --> 99:59:59.999
that breath analysis can provide,

99:59:59.999 --> 99:59:59.999
even for people that are not
part-time hypochondriacs like me.

99:59:59.999 --> 99:59:59.999
Thank you.

99:59:59.999 --> 99:59:59.999
(Applause)