Player is loading...

Embed

Embed code

Transcriptions

Note: this content has been automatically generated.
00:00:01
so hi to everybody in recent years we're seeing more and more technological changes
00:00:07
and probably as most of us think many more will expect us into the future
00:00:12
what's the changes within boulders well rehabilitation which is
00:00:16
probably probably undergoing real revolution into next year's
00:00:21
first of all when my name is never that's already have a mustering fees except eugene information engineering
00:00:28
i'm senior researcher here at assures us over like what i work um biomedical data
00:00:32
analysis and rehabilitation i'm part of the yells you wanted and automatically script
00:00:39
which can't in total twenty twenty five persons working on the development of artificial intelligence
00:00:46
and machine learning and keep learning procedures for biomedical data now this so if you are interested to know
00:00:52
more about our group or to contact that's just come back here technical or drop us an email
00:00:58
so in my presentation i will uh first explain what we uh think what
00:01:04
uh does every mutation revolution me why we should expected
00:01:09
and what should we expect from the rehabilitation recognition
00:01:13
finally i would describe it how our group is contributing to you need does it
00:01:18
so let's start from what we mean as rehabilitation revolution
00:01:24
well someone said the twenty first century will be the sentry of the robotics revolution but is it true
00:01:31
well percent fiction for sure we've seen many different movies involving a rowboat and race
00:01:38
human it's of any kind and shape according to some recent results well
00:01:43
maybe yes this video basically us seeing atlas one for
00:01:48
the robots delphi most the place wasn't in
00:01:50
any which is actually capable to do things that i'm not at all people to do
00:01:57
however the results can it's a a change of each
00:02:01
and the quality when the situation start changing
00:02:05
basically a robust to not react well to
00:02:09
situations that have not been predicted before
00:02:13
so anyway it is foreseen that the robotics revolution we'll
00:02:18
extend to wearable rehabilitative and assisted robotics as well
00:02:23
including prosthetics and for instance access skeletons for rehabilitation
00:02:28
or assistance in the directory seen here
00:02:31
it's in one of the work resulted i think is most fascinating and it is
00:02:35
a bilateral amputee that is controlling a robotic and processes using targeted master innovation
00:02:41
the project cos it quite a bit of meat dollars to united states but the results are astonishing i would say
00:02:50
so why should we expect a rehabilitation revolution
00:02:55
well for three reasons first because it is needed
00:03:00
second because it can allow to create profit for companies and ferret
00:03:05
because the reason you put you needed to do weights nowadays
00:03:10
so for some reason why do we need a rehabilitation revolutions well
00:03:15
patients uh needed assistance and limitations are constantly augmenting us
00:03:20
was pointed out in several presentations before together with healthcare costs
00:03:25
nowadays we have two point wine put two point one median amputees
00:03:30
in due west that cost approximately twelve billion dollars each year
00:03:34
and the number of amputees it's expected to augment in two thousand and fifty two three point six meetings
00:03:42
currently we have forty seven point eight million of people us of age sixty five and
00:03:47
over and the number is expected to almost double again in two thousand and fifty
00:03:53
as well we have one hundred and seventeen me then of people with chronic diseases
00:03:58
as well also this number is expected almost to double in
00:04:01
seven fifty so all the situation require the development
00:04:06
of a see steve and rehabilitative techniques that can be affordable in the future and that can help
00:04:12
governments and he trends to a soul these social
00:04:16
problems related to uh people with such impressive
00:04:21
why uh rehabilitation evolution can allow companies to create a
00:04:26
profit but nowadays research and development in rehabilitation
00:04:32
uh require huge investments to the companies for prototyping
00:04:36
for testing for complying with the low requirements
00:04:40
the market in particular in the field of rehabilitation despite
00:04:43
his beak evolution being the slide before it's
00:04:46
not sufficiently big to absorb big investments so good
00:04:52
products lead to very very high cost
00:04:55
as you can see in this picture wait nowadays has several different hands that are they advanced
00:05:02
and it can put on the phone different movement however the cost a lot the average causes
00:05:07
are over fifty thousand dollars which is something that cannot be afforded by hand amputees
00:05:12
as well as the skeletons for for instance per plate you people the easily but usually cost
00:05:17
over one hundred thousand dollar considering just the keypad without installation the adaptation to the patient
00:05:23
and the price with these can increase over two hundred fifty thousand so better products and lower
00:05:28
costs can allow to reach more patience to increase the market and to develop new business
00:05:36
why they're easy opportunity every every pollution nobody's well because
00:05:40
modern technologies can make development easier faster and cheaper
00:05:46
their sin before uh uh about a a three d.
00:05:49
modelling additive manufacturing and rapid prototyping so pretty printing
00:05:54
open source software and at the feature intelligence all of those techniques and
00:05:57
eight nowadays quite easily available and they can allow to reduce prototyping
00:06:02
cost reduce testing calls and to comply with low requirements actually only when
00:06:06
it's really necessary so when the product gets to the market
00:06:13
so let's describe it more in detail that the uh such technological uh uh um opportunities
00:06:20
first one three d. modelling what these three d. modelling it's the process
00:06:23
of developing mathematical representations of objects in three dimensions yes specialised software
00:06:30
so nobody is the result growing up resource community what why the design three d. models
00:06:36
and release them on a specific websites from which
00:06:40
uh users can download them they can buy them and reuse them or adapted to their needs
00:06:47
um in one interesting thing is the design is not made only
00:06:52
by professionals but also bought by many people including the users
00:06:57
so for instance here you see a web page that allows to share designs here you see
00:07:03
a three d. model of a three d. printable prosthetic and and
00:07:07
here you see probably the coolest things all web site that
00:07:10
actually really easy is specific three d. models of prosody cans for children that he is that are particularly uh these
00:07:17
a difficult to target problem in prosthetics because the the grow
00:07:22
and so they need to proxies to be adapted quickly
00:07:27
second the uh technology factor that can uh uh um foster datum
00:07:32
of the of the rehabilitation revolution additive manufacturing and rapid prototyping
00:07:37
what are those well that that process using which materials are joined together
00:07:42
under computer control in order to create a three dimensional objects so from the
00:07:46
designs on a and the design models we can reach real products
00:07:51
nowadays as was said in no one of the previous presentations we have many materials get can be three d.
00:07:56
printed including polymers composites metals and biological t. should as
00:08:01
well and applications range by industry to domestic use
00:08:05
medical use healthcare foot human cell production and rehabilitation as well
00:08:11
so here you see a set of nowadays available pretty printed prosthetic and that people can download
00:08:18
from the model for for which people can download from the web the models and can print by themselves or to use them
00:08:25
third and last factor is open sort of software and the artificial intelligence
00:08:31
so what is it obvious software with source code that is released under licenses that grants users
00:08:38
just started changing distribute the software so that i think the p. software that people can download ended up to their needs
00:08:45
uh no what he's the main feature is that there are many different
00:08:48
software that can be applied to vary from the mains including robotics
00:08:52
for instance the ross system robotic operating system is probably the
00:08:56
most diffuse a framework to develop robotics environments worldwide
00:09:01
as well as there are many different frameworks for artificial intelligence for developing for instance the
00:09:07
peril networks to study images or time series of uh or any kind of problem
00:09:14
what should we expect to buy the uh um imitation revolution
00:09:20
well we should respect that scientific research and
00:09:22
companies can benefit of those available resources
00:09:27
it can develop innovative and more advanced technology and products and finally reduce costs
00:09:33
that's producing benefits for the patients benefits as income for the companies
00:09:37
and benefits for the health care system that we as we have seen
00:09:41
before it's gonna have some problems in the next twenty thirty years
00:09:46
so how is our group enabling the rehabilitation revolution so how are we participating to eat
00:09:53
well uh in two thousand and fourteen we really is something i'm pro which
00:09:57
is a publicly available database for surveys selector my your three data
00:10:02
that uh currently includes seven data sets more than one hundred and twenty subjects including thirteen amputees
00:10:08
and it has over five hundred users worldwide actually i would say quite a bit over five hundred
00:10:14
what does electronic your fee mean well first of all you need to know that actually the muscles that control
00:10:19
our hands are located here the one that extend the fingers here the one the flex the fingers
00:10:25
so people that are amputated below the elbow they still had the muscles that used to control the hand
00:10:30
what we do is we place sensors on this uh phase of
00:10:34
the scheme of both in text objects and hand amputees
00:10:38
and we record the activity of the muscles into the stamp or into before
00:10:45
as well on it that's subjects we put a lot with sensors that can record the
00:10:50
kinetic movements of the fingers so that sense of with lex when each finger flex
00:10:55
we record the data when the subjects repeat a set of approximately fifty
00:11:00
different of movements and we we use them on the web
00:11:06
the analysis of such data allowed to obtain until now very interesting results
00:11:10
for instance uh it allowed to um underlies the
00:11:14
uh effect of clinical parameters on arc ability
00:11:17
to classify different kind movements in particular length on the stamp with so that is
00:11:22
um related to our capability recognise the movement as well as phantom the sensation that
00:11:27
is the sensation to have the and steal the are attached to the stand
00:11:32
it allowed us to develop a numerical models for grasping
00:11:36
that that can allow to develop better robotic hands
00:11:40
that can allow to develop better prosthetic hands as well that can allow to link
00:11:45
robotics prosthetics and physiology finding links between hand synergies and uh such models
00:11:53
currently we're releasing publicly available software so uh we planned release pilot data
00:11:58
analysts creates this summer we plan to release the data acquisition software
00:12:02
by the end of the southern in eighteen and every other time working model control software in two thousand and nineteen
00:12:10
there's also to show that actually be which devices that cost
00:12:13
around three hundred years like this one can get results
00:12:18
as comparable to be a real expensive devices that can cause eighteen thousand euros or more uses
00:12:24
standardise bitterness public showing that actually good results can be obtained with very cheap prices
00:12:32
so after these we tested such data acquisition environments on hand them to to use
00:12:37
doing these movements so the face these kind of movements this kind of movement and these kind of moments
00:12:43
and outside you can see a day when i can describe this in more detail to you
00:12:46
results show that entities are capable to perform such movements pretty well
00:12:52
we applied the specification system to the
00:12:55
control of augmented reality uh the
00:12:59
actual hands so here you can see a francesca one of assistance
00:13:03
that he's using the control system to control and to grasp at the train to grasp
00:13:08
different objects in oregon augmented reality using microsoft holland's
00:13:16
as well we developed a a three d. printed hands process models
00:13:21
starting from philly mother that was released by diseases cannot
00:13:26
this handy since right it's s. census and timothy senses and as you can
00:13:30
see that's these saying is it is it works seem to work
00:13:36
we developed a set of sockets because once okay you have the hand as you can see outside but then you need to use it
00:13:43
we developed a sort of stuff except that allowed to uh index objects to use the hand as well as
00:13:49
that can allow had amputated subjects with different level of computation to use the and easily and fast
00:13:56
as well we develop more than sockets like this one that abate instead of three d. reconstruction of before
00:14:03
we and allies day man dynamic and climatic processes
00:14:07
all the hand computer we need also during
00:14:10
a while you're in the execution of different and crest with the real hand with
00:14:15
a real handing odyssey is it good is not good and how can we
00:14:19
improve case it was also basically summarised into these video that actually we submitted wannabe
00:14:27
remain relatively conferences worldwide and uh because right so we thought of that
00:14:33
the uh we performed and we didn't really really land on a several
00:15:20
well as you can see by the smile on the face of the subjects the
00:15:24
the results were pretty good they found very easy to use the hand
00:15:29
it took like half an hour to get use to use it
00:15:32
while normal uh training procedures take up to two weeks
00:15:37
and they we're really impressed by the the the how easy it was and by how cheap it was
00:15:42
because finally think passive that cost that around that's it's less than one thousand francs for sure
00:15:50
currently we are as well um it trying to explore it
00:15:54
eye hand coordination in order to improve our control system
00:15:59
basically a hand coordination is the phenomenon for which we look at the objects before resting them
00:16:05
so we basically placed happier of classes with eye tracking on the subject and
00:16:10
we are recording them and where they look on to the objects
00:16:15
before performing grass we hope to use those multimodal data in combination with electra my your fee
00:16:21
and computer vision noted to make process that the market more autonomous incapable to understand
00:16:26
if for instance want to grasp the bottle from the side in these two eight or from the top in this way in order to remove the cat
00:16:35
so in conclusion take a message to rehabilitation revolution
00:16:39
use near i think thanks to technological advancements
00:16:44
companies can benefit of it by developing new business models that
00:16:48
are more that the current economy and scientific trends
00:16:52
d. e. l. c. unit and they might give true that are based here and there are
00:16:57
open to collaborate with you would experience impassioned if you plan to join the rehabilitation revolution
00:17:03
so thanks a lot to uh our group in particular to professor annual or that is the chief of the group
00:17:09
mitochondria activities the p. h. d. student and some working on many of the things that i presented now
00:17:15
vegetable or more that developed the uh augmented reality systems are
00:17:19
technical c. stands all the collaborators what wide and
00:17:23
most of all actually can we say to the financing and it is never ever
00:17:26
helping us to perform these antiwar groups obviously that as you can see people
00:17:37
thank you that's a group because children some
00:17:47
it's question
00:17:56
i'm sorry i was just i was just going to ask i think you mentioned the cost of one um yep
00:18:03
can you just a repeat example a uh yeah we of what the what the hell did we
00:18:09
developed for the commercial once the commercial one over fifty thousand years oh okay thank you
00:18:18
that's true is one let's say the one that can perform that can move each finger separately then there are
00:18:23
handed turn just close and open and those cost much less and around a seven thousand us ten thousand
00:18:36
um we we have seen the the prototypes that you one present in your own suite in the
00:18:43
room b. one does it very presume home use the the big full those products to be
00:18:49
why do you commercial you with people at this low price that you mentioned what is do you think the
00:18:56
path of the route map do only that you have to to to to to the cool thing or
00:19:03
what i what what what i presented and which was not easy actually to to to explain it but ease that uh
00:19:09
the field isn't fun to sing much faster than uh than
00:19:13
the markets and then what a single company can do
00:19:17
i mean uh probably now there are several people worldwide that are designing the romance
00:19:24
and that are testing you solutions and they basically don't need
00:19:27
certifications they don't need a um a cow that
00:19:31
protocol approvals so the the feel is usually growing extremely fast
00:19:37
it q. fast in the last like five years
00:19:40
do in a um they are like timothy
00:19:43
cans are nowadays available so that can
00:19:46
be controlled like you do we doing weird movements of it with our
00:19:51
or uh they're also u. n. g. control prosody cans that are even already uh
00:19:55
already now available for more complex sense so what we're basically trying to do
00:20:01
i hope it will be soon like let's see in a five
00:20:04
years three five years i would have to be realistic yes
00:20:10
okay
00:20:11
because there's or to press the though 'cause you know it could double as you with a very sensibly

Share this talk: 


Conference program

Mots de bienvenue
Sébastien Mabillard, CEO | Swiss Digital Health
15 June 2018 · 9:07 a.m.
Mots de bienvenue
Victor Fournier, Chef de Service de la santé publique | Canton du Valais
15 June 2018 · 9:11 a.m.
Mots de bienvenue
Jean-Albert Ferrez, Président | Fondation The Ark
15 June 2018 · 9:19 a.m.
Mots de bienvenue
Laurent Sciboz, directeur Institut Informatique de Gestion | HES-SO Valais/Wallis
15 June 2018 · 9:24 a.m.
L'écosystème d'innovation ouverte du CHU Sainte-Justine: une grande communauté gagnante!
Kathy Malas, resp. de la Plateforme de l’innovation et des Fonctions des maladies chroniques et aiguës | CHU SAINTE-JUSTINE (Canada)
15 June 2018 · 9:33 a.m.
Les livraisons par drones : vers une amélioration de la logistique dans le domaine médical
Janick Mischler, Program Manager | LA POSTE SUISSE
15 June 2018 · 10:07 a.m.
E-health et intégration des soins
Marc Cikes, CEO | MEDBASE ROMANDIE (Suisse)
15 June 2018 · 10:32 a.m.
Questions réponses
Remi Gauchoux, Business Development Director - Carenity
15 June 2018 · 11:46 a.m.
Futur de la santé mobile
Dr. med. Patricia Sigam, CEO & Co-founder, digital Med-Lab
15 June 2018 · 11:54 a.m.
Democratizing Data-Driven Medicine
Tarik Dlala, VP Marketing, Sophia Genetics
15 June 2018 · 12:14 p.m.
Ada inside
Vincent Zimmer, Ada Health, Berlin
15 June 2018 · 12:39 p.m.
biospectal, the optical revolution in hypertension monitoring
Prof. Patrick Schoettker, CMO. Biospectal
15 June 2018 · 2:06 p.m.
3D Printed Medicines: A Digital Pharmacy Era
Sarah Trenfield, MPharm, Senior Formulation Scientist, FabRx Ltd.
15 June 2018 · 2:23 p.m.
The Digipharm experience
Ahmed Abdullah, CEO & co-founder, Digipharm, Basel
15 June 2018 · 2:45 p.m.
Enabling the rehabilitative revolution
Dr. Manfredo Atzori, HES-SO Valais Wallis
15 June 2018 · 3 p.m.
Team Gamified Multi-sensory Stroke Rehab
Jean-Luc Turlan et J-P. Ghobril, Lauréats Arkathon 2018
15 June 2018 · 3:22 p.m.
Secure and Trustable EMR Sharing using Blockchain: Open Challenges and Lessons Learned
Alevtina Dubovitskaya, HES-SO Valais-Wallis
15 June 2018 · 3:38 p.m.
Conclusions
Sébastien Mabillard, CEO | Swiss Digital Health
15 June 2018 · 3:59 p.m.

Recommended talks

Mots d'introduction des Instituts de la HES-SO Valais-Wallis
Laurent Sciboz , Directeur des instituts informatiques
3 June 2016 · 9:26 a.m.