Player is loading...

Embed

Embed code

Transcriptions

Note: this content has been automatically generated.
00:00:00
it was very difficult for us to pick some interesting topic
00:00:04
for for the state because the over interesting and you know what we're discovering everything so we could you
00:00:10
spend days and days going into detail and so and so what we've been trying to do is
00:00:16
excel on that we just uh like first allies some knowledge for you to better as
00:00:20
the stand the future and we try to make it progresses through the day
00:00:25
uh so hopefully you will like it and if you don't we try to make it better
00:00:29
for the next line and anyway so feedback would be important now i've we said before
00:00:35
uh is sign takes a scientist medical doctors finish
00:00:40
in researchers so they can we're all
00:00:43
and that's very important that we all interacting together because if we developing all solution in the corner
00:00:49
you know we have like as many solution that was people because we won't match
00:00:53
so networking is very important discussing problematic to don't be afraid to ask question
00:00:59
uh don't be afraid to interact and uh would make the best so before any
00:01:05
further uh going uh to go further sorry uh i'd like to introduce uh
00:01:11
what does it domestically so they had as the head of the remote
00:01:14
allergy and uh get the idea of these uh exciting day
00:01:18
so is mortar printer eighties mine rather than the recognition is always go for it
00:01:24
uh but he's also good clinicians all those don't worry about that but you try to apply to determine metallurgy and a
00:01:30
and a lot of things to to the to the patient and that's why we're so have some patients you sell
00:01:36
no further worked it which ought to be nice defied a bit somewhere that's a lot of people
00:01:41
mixing up terms so that would be the first that all along well thank you very much
00:01:47
and thank you d. d. and um we would like to um thank the sponsors
00:01:51
obviously uh we could not um afford it uh you know to pay
00:01:55
for this venue to be yeah and pay the two now so thank you
00:01:59
for all um industry support to make this possible and and actually
00:02:04
to share ideas also and to to for um to to to follow to
00:02:08
find solutions together so with that i would like to um start with
00:02:13
the first talk so that's not an easy talk right it's a very
00:02:18
large field team is suffocation of the digital health work so um
00:02:23
first of all you might ask yourself why are we here today why are they organise thing this well
00:02:30
obviously i'm digital solutions are moving extremely quickly fall but and this um has implications
00:02:37
on patient care and always daily life as doctors that that's for sure
00:02:41
uh it's not in the mid medical curriculum so um mostly then no set courses upcoming
00:02:47
but still we not trained in a i for example when we study medicine
00:02:52
for example um maybe we should do so uh we we can discuss this later
00:02:56
and there is a lack of interaction between and tech and
00:03:00
data scientists and doctors and industry so this is the
00:03:05
this is why we choose and one day especially for this
00:03:08
and personally this is my um most important um
00:03:12
reason to do this and i haven't as a scientific in and research interest
00:03:17
uh in this field right because our role in university hospital is too
00:03:22
you know uh to create knowledge and and and to push the feel for it so
00:03:26
and and and yes there are sessions in you'll uh for example i see are usually it's a small
00:03:32
session and and i can not not the full day right we have time time and also to network
00:03:38
so uh this on my conflict of interests i have no teachable conflict of interest if you wish
00:03:45
i'm just uh are you sure you had no digital conflict
00:03:50
of interest uh i look so uh you here um
00:03:54
uh well no actually i do uh i do have a digital contagious and human first of all so you
00:04:01
know i'm interested in uh that that technology it helps us right and and of course the patient
00:04:08
i'm half of my family are programmers so um we
00:04:11
do work we have projects in sun ten years
00:04:15
and never in the academic setting and now this is now possible with this position here
00:04:20
yeah it was and there's one ongoing and deepening project with units of what's
00:04:24
to fry but so we will hear million grimaces uh in the afternoon
00:04:28
i'm talking about this so yes we talk about this when we meet sunday for lunch
00:04:33
and yes i also um trading it digital and academic case but from because i think
00:04:39
we should more discuss cases interactively and maybe we can uh use something a shareware
00:04:46
that and and um hospital room touches can use to discuss and very interesting cases
00:04:51
and yes um is it implied in clinical medicine we started it for our chronic
00:04:57
pain patients they have now this watches the smart watches which you haven't been
00:05:03
and actually we use an end to an allowance um sleep so as
00:05:08
we know that in a chronic pain sleep is very important and
00:05:12
a a very often the that problems so with this uh
00:05:16
watch uh with the software we can say exactly which
00:05:20
a kind of sleep faces the patient is in here he's you know it
00:05:24
he says uh i don't sleep at all but he's like a stone right
00:05:28
unless he's eight hours in bad without moving but that's not possible
00:05:32
awake so uh we use this now starting from scratch
00:05:37
a digital a station and i'm yes it started with
00:05:40
the computers second world war um and and
00:05:45
you know computing data data processing obviously it first
00:05:48
revolution then second revolution the internet brought together
00:05:53
computers so they were connected we could communicate better i increased to computing power
00:05:59
then actually i'm smart phones became computers so that introduced mobility
00:06:05
right so now um we're already there we can have it
00:06:09
all connected than a bit the the full gotten uh
00:06:13
and technology i in my opinion the ah it's censoring so it's amazing what
00:06:18
new sensors what cameras okay but also at the kind of sensors are
00:06:23
i'm about to be developed we have artificial noses right now right that
00:06:28
can detect a key to us it is is in diabetic patients
00:06:32
um even cancer there there are tests right can they the the computer
00:06:37
the artificial knows us now that you press changes when you
00:06:40
have a certain forms of cancer could be new phone for example
00:06:44
so this next a generation censoring is is very interesting
00:06:48
and what's new it's uh well we talk not into that of things so uh your fridge
00:06:53
or worse um channel makes will tell you what you have to um by um tonight
00:06:59
and um uh automatic driving in flying i laugh this topic
00:07:04
i think it's more than playing around so you
00:07:06
might think i'm not a carton that car is something
00:07:11
you know not necessary but somebody with epilepsy
00:07:14
uh or elderly people or drunk people you know this could be very useful
00:07:19
uh in all sorts of situations and i think that will come
00:07:22
and then one began topic for today's active artificial intelligence in the broader sense
00:07:27
right to so no to the machine learning we will talk about this
00:07:31
um less so about robots artificial intelligence maybe to start with
00:07:36
i'm not afraid of artificial intelligence i think um artificial intelligence will help us
00:07:43
will assist us and i think it will never replace a doctor
00:07:47
i think it will replace certain jobs like uh the taxi uh a central
00:07:52
where you call you know this is gonna change like other um
00:07:57
chops um disappeared um in the past this will be the same but
00:08:01
i'm definitely not doctor so i think artificial intelligence will assist us
00:08:06
uh uh like the servants so it's a new kind of revolution after the industrial revolution
00:08:13
well our people did not work hard at work hard in the factories anymore we had machines now we have
00:08:20
i am a i to help us to assist does and you know for
00:08:23
an affordable price let's put it this way if you need new
00:08:27
trousers uh you have nobody uh you alone well you could maybe first
00:08:32
ask uh the machine well you know what i like uh
00:08:36
and you know some uh offers a in town can you suggest something so uh
00:08:41
and the same is for for madison i think it will assist does
00:08:44
not more not less but um i think that's good and ideally like a
00:08:49
servant uh it's would be reliable discrete and confidential and an application
00:08:56
so when we talk about um digital solutions and you
00:09:00
might um you know divide this into through we
00:09:05
um parts first on the user interfaces so this can be you
00:09:09
french if you wish um uh in can be your car
00:09:14
but also a more important in health you have for some patients reap
00:09:17
reported outcome your mobile that's the that's the user interface well
00:09:22
all we put data uh you know the as you can data we put it in the computer than
00:09:27
almost more importantly you have the um the data management platform and and so somewhere you
00:09:34
have to store this data and it's important where that regulations especially for self
00:09:39
you know you can that you cannot uh uh save this on the computer the server
00:09:44
uh whenever you want this should be clear uh in house server and that's important
00:09:49
so the role of this data management that is to collect the data
00:09:54
that's very very important uh to organise it to store to share
00:09:59
and to provide provide them and safety so that there's no attack if you wish but also privacy
00:10:05
so and this is getting more important puerto so which role for some should
00:10:10
a patient have to say uh who has access to my data
00:10:15
so right now we cannot decide but maybe in future and the applications there are companies also in in in those on
00:10:22
and who who work on this and so uh i think
00:10:26
data privacy is a big issue and obviously compliance
00:10:29
with the new regulations coming up uh all over the world
00:10:34
um be stated last one has to be complied
00:10:37
and then as a thirds that we have artificial intelligence it's there already
00:10:42
um which an ally says low instances as as we said uh it
00:10:46
is important to understand complex patterns and we'll file way to understand
00:10:51
all uh diseases in madison right it's often that we you know we we name we put a diagnosis but we
00:10:58
really it's we think it's that's a strange thing so we
00:11:01
need help uh with with complex situations also noticing
00:11:05
and it can help us to reduce um at the states because
00:11:09
we do mistakes and patients time because of mistakes in madison
00:11:13
so why not having a i to help us to reduce uh this misstates let's be honest but it says
00:11:20
and if this saves time and money i mean even better because by that
00:11:24
time is an issue we have less and less time to uh also
00:11:28
from regulations to stay with the patients to talk so we would need a
00:11:32
um solutions that allows us to have more time to the page
00:11:37
so this is what we're going what are the limitations well when a
00:11:41
wal mart um it's had has paid into the a card
00:11:45
that way it's your temperature and pulse of of the you know when you go shopping
00:11:50
add this information to i don't know if i you know if i'm
00:11:53
happy to give this information right so this could be a limit
00:11:57
right so uh uh and and they would see that um it's
00:12:01
certain drinks you know uh i'm i'm interested and sweating
00:12:05
and then i would get in a edward from uh this company later on
00:12:09
so this would be a limitation but we have to talk about this especially in health where
00:12:14
data is very um you know uh even more difficult and and and to discuss then
00:12:20
uh is this true or are we they already um let's put the uh the numbers for mobile health
00:12:27
so they're the revenue the billion dollar a turnover and
00:12:32
crew from two point four billion billion dollar
00:12:36
and to twenty three billion dollar two thousand eighty that a massive marked so here you can see that this
00:12:42
innovation but this also in need uh for this innovation so we're not talking about playing around this is
00:12:49
this is a real so people spend money on this now going but what is mobile health well
00:12:55
apple obviously spent a lot of a lot of energy into health so you
00:12:59
have the easy she now it's very interesting you can uh uh and
00:13:04
measure you pluck pressure uh we're we're we're bills we will talk about this e. diet noses
00:13:10
uh almost everybody now uses a two or other platforms to have information
00:13:16
uh when when there's a certain problem we use this is monitoring page report outcome
00:13:21
digital by market will come i hope so if i could choose between
00:13:25
uh n. a. and plus sample uh for a c. r. p.
00:13:29
and a special kind of photo full already which that tells me i'm inflamed or not i would probably uh
00:13:36
first use um uh at the ford or or or another kind of sensor uh to do this
00:13:42
and we what is digital therapy so we we have
00:13:46
this already but it's goes more into behavioural intervention
00:13:50
so especially uh for arm your health uh issues in and
00:13:55
psychological issues depression but also it's used for chronic pain
00:14:01
what to do to relax and how to deal with pain but also interesting for compliance
00:14:07
but but um you know uh increases compliance could start with a little
00:14:10
a reminder being that you take your draw right that's interesting
00:14:14
and and maybe even to fort and for three to target and
00:14:17
to make sure that for example for for a count
00:14:23
but you it reminds you that you you read as it should be in one of three hundred sixty for for example
00:14:28
and then add tele diagnose is uh probably coming more you know um
00:14:33
you have to see the patient disk is coming up and again artificial
00:14:36
intelligence and i think there's an effect can become the patient and
00:14:42
um my phone tells me how will i be next week or next month
00:14:48
this could be interesting right because i'm planning a trip to barcelona for example
00:14:52
so um if a oh i could tell us or give us information paste
00:14:57
on genetic background one inflammation et cetera on my
00:15:01
profile ah how on a walking around
00:15:04
this could be interesting right and then uh and this is obviously a
00:15:08
and that extreme and uh but i hope we get there um
00:15:13
and i i assisted treatment so it will become your will help us
00:15:16
to choose the right treatment right it will not designed obviously but
00:15:20
it will it will help us if the computer has enough information to predict uh uh the right one
00:15:27
and it's a it's a kind of digital circle a a
00:15:30
psycho babble which is about to start and again um
00:15:35
this will change the roles of everybody of the patients a cheap b. and specialists
00:15:41
and i'm looking forward to this uh obviously there's regulation uh required it's a big uh
00:15:47
a topic the uh we cannot do what we want right everybody
00:15:51
it's like a a shot at an early uh with tracks people cannot
00:15:55
just say you know we have a new drug that's wonderful
00:15:58
you have to show and prove that it's wonderful and that it's safe and the same will be true for digital applications
00:16:05
and then we by the way that's also an interesting point we can use hyper personalisation
00:16:10
so it's as i uh look uh next week you are you are in
00:16:14
good shape your you uh array is is come the weather is fine
00:16:18
actually you can arise is good i do some sport right so hyper position is is also an interesting thing
00:16:25
now it's is already in place i'm actively intelligence diagnostics clear
00:16:30
yes we have this for that it written about the
00:16:33
a digital pathologies is coming and and also um
00:16:37
in cardiology obviously so we are already here
00:16:41
uh in other fields not yet invented comedy or chronic inflammation i hope this will come about therapy
00:16:47
uh this has been published uh recently in sepsis
00:16:51
so by um and machine learning with all
00:16:54
the information you get from the patient and this is massive in in intensive care
00:16:59
uh it was better to um choose the right treatment right the
00:17:03
machine but the interesting part is that the best results were
00:17:07
obtained when machine and the doctor agreed so so this is
00:17:11
interesting and it's real now again regulation there's um eh
00:17:17
regulation i'm coming up like the preserve program from the f. t. a.
00:17:22
and what is interesting here it's a bit different compared to uh and the medical devices for example if you will have
00:17:28
a new medical device you want to register then uh and you well they they they look at the device
00:17:35
and here uh with um a digital solutions they look a more on the
00:17:40
provider be develop right the company uh if they are able to do this if they have
00:17:46
quality control measurements et cetera and then they look at the software as a medical device
00:17:51
okay so this is a bit different thing compared to medical devices on trucks
00:17:56
um but it's very uh a very interesting and what is coming up
00:18:01
so time wise uh we still have some time i we're
00:18:06
yeah just this is what mccain say things um about this whole thing
00:18:10
and so in external view as so they and and i agree actually so we can
00:18:15
probably a bit better and diagnoses um we we can also with
00:18:20
electronic patient record i hope it comes minimise costs improves outcome
00:18:26
um the patient will become more independent less paperwork that's a good thing come more time for
00:18:31
patients right for the patients um they can
00:18:35
better communicating future um it's it's um
00:18:41
you know on the page has more data he's not just coming saying look on time have pain here you will
00:18:47
uh provide with his symptoms also let's say an evidence by data to the
00:18:53
doctors so we will be more independent and has more to offer
00:18:57
uh if if you wish that we can then discuss by the way
00:19:00
already now five percent of all two clicks on health related
00:19:04
that's a that's a lot like and seventy five percent of you compilations um go online for health
00:19:09
information probably it's even more now and there's some study showing that you have better compliance for example in in in c. o. p. d.
00:19:16
if so uh this is has consequences again that
00:19:21
patient would be more independent um he will take more control on the disease in future
00:19:27
um he will be empowered by by this information and then they are might you know
00:19:34
try different the uh products because they have more information and and share information
00:19:39
uh and this is interesting about now just a a
00:19:43
little example and of the limitations so if your i'm type into google a a symptom
00:19:50
what you get is information right about cough if you type in
00:19:56
cough and fever what you get is a direct those it's
00:20:00
right which could be trunk items so if you do this well you you get on on these
00:20:04
sites it's not crudely itself yet other sides and they say how these two things are
00:20:10
uh he's he's been quite as well this you know to taking on this and this
00:20:15
okay they also a clinic so do this uh even the aquatic in one channel which is uh
00:20:21
ah journal for patients in german one they do this so that's a no
00:20:26
brainer that seems very easy ride algorithm is easy but this one so
00:20:31
but now if we have other factors chills age you
00:20:35
know it gets more complicated you have more data
00:20:38
so can the computer still take the right decision well f. if you has all this information yes
00:20:44
but it's more difficult business pomona left the hospital lines and stuff so it's getting more interesting
00:20:51
patient desire but you needn't you need this information to great now person and with
00:20:55
machine learning you probably you you can handle this and take the right decision
00:21:00
but let's let's see this limitation you have two persons well let's say
00:21:03
it's the same person one one of the two aspects syria he's
00:21:06
been at it for two hours so who would say who is sick what do you think left or who is for left
00:21:17
always for right
00:21:20
okay exactly right so it's just you know a
00:21:24
bit pale the lips it's very discreet
00:21:27
so this would be a two hour so this would be very very difficult for machine to find the right diagnosis
00:21:33
and do the right things right so this page you you have to see and sometimes it's exactly intuition
00:21:39
right you could say well we have is genetic background and he
00:21:43
is at risk for sepsis so maybe this will come up
00:21:47
but but for the moment it's good to see the patient right so there will be shift
00:21:51
um i am off of a consultations with self diagnosis in for example skin cancer
00:21:57
mortality likenesses i think the role of the g. p. will be more important in theory he could
00:22:03
check adapted with you know with the with the machine why going to an often with just one and there would still be the
00:22:10
specialist and and healthcare professionals then uh to do hands on work
00:22:13
and and and take decisions okay so to sum up um
00:22:19
where i see a n. s. a. and a big opportunity is
00:22:23
to put or data that we create and this will be more
00:22:27
and more data together because it's too complex for us
00:22:31
yeah we need machine learning i'm too great algorithms to help us to
00:22:35
bring this all together as part of an integrated data platform
00:22:39
as it is already done um in oncology for example i hope this will come to chronic inflammation
00:22:45
um that this is we have especially with dynamic data but also
00:22:50
let's say economic data off the um tissue like they do with the cancer
00:22:55
right now so you need generic information about the host if you wish
00:23:00
uh even about the my probably on that but also the inflamed tissue
00:23:04
and and with that we might have a more information which treatment
00:23:09
um to use it's still the question who takes decision
00:23:14
so it's nice to have such a data platform with all this at
00:23:18
the end somebody has to say okay we do this or this
00:23:21
and often it's you're not alone we also have a dermatologist for example gas
00:23:26
intros do we need to interact so it's platform should allow also communication
00:23:31
and then uh you know maybe one day we will have companion use also in mythology that means
00:23:37
one finding here means and not as clear treatment there that
00:23:41
that's company news so in oncology there there and
00:23:44
i hope uh we can use this also for three to target and and and to increase compliance
00:23:50
so we there somethings to discussed we can do this later
00:23:53
how will this be implemented um what's the scale
00:23:57
well in our field for for chronic inflammation data privacy can we ask for data donation
00:24:05
right some people would say i will like to give you my data so data donation could be an issue
00:24:10
does a i can create a evidence or do
00:24:14
we still need perspective uh i'm ed trials
00:24:18
uh do we need plus a bleak future or can we use real world data uh for a
00:24:23
prospective trials so people you know so we don't have to give people a placebo anymore
00:24:28
and and can go it also indices prevention how can we avoid data
00:24:33
so and this is it uh ladies and gentleman a a short overview about digital half i thank you for your attention
00:24:43
whoa
00:24:46
ah so you can clearly see some two z. s. overlay
00:24:51
on the time but that's a that's a mass
00:24:54
in the song to z. at that but a lot of energy into it that that's great for
00:24:58
for for the departments all thank you to last so uh we said the question for later because we're running a week
00:25:05
later ready uh but uh we have a form so would be able to discuss later also was complete global introduction
00:25:11
and then we start really to go into it no always a remote
00:25:15
allergies so it's uh i'm very pleased to introduce the next session
00:25:18
which use a a detainee eat a few years and possible solution re
00:25:23
metallurgy and you will see in that session we have to
00:25:26
uh presenters some extent possible he worker from basil will as a remote allergy
00:25:31
we're really talking to the need the fees for remote elegy standpoint
00:25:36
and then we'll likely to also have a a a yeah nice cosmo off which
00:25:41
is a patients uh but it turned out to be also when at specialist
00:25:47
so we see uh how you will react to yeah uh
00:25:50
to a postal worker presentation and also chaise on experience

Share this talk: 


Conference program

Demystification the digital health world
Thomas Hügle
1 Feb. 2019 · 9:41 a.m.
249 views
Digital lessons learned from musculoskeletal radiology
Patrick Omoumi
1 Feb. 2019 · 11:37 a.m.