driven by new trends in artificial intelligence: robots in ... · pdf filedriven by new trends...
TRANSCRIPT
www.ima-zlw-ifu.rwth-aachen.de
Driven by New Trends in Artificial Intelligence:Robots in Medical Applications
Leonardo-Modul „Robotik in der Medizin“ RWTH Aachen University
Aachen, January 18th, 2017
Univ.-Prof. Dr. rer. nat. Sabina Jeschke
Cybernetics Lab IMA/ZLW & IfU
Faculty of Mechanical Engineering
RWTH Aachen University
2
18.01.2017
S. Jeschke
Outline
I. Introduction
Da Vinci as a starting point
towards a more general approach to robots in medicine
II. Physical robots
in different application areas
in different shapes and design principles
III. Non-physical robots
in a nutshell
from Google Flu to Watson and beyond
IV. Summary and Outlook
the question of a creative artificial mind
4.0 trends in labor and employment
3
18.01.2017
S. Jeschke
Introduction
Da Vinci is only the starting point…
!
The most prominent medical robot today: the “da Vinci® Surgical System” for laparoscopy
Designed by the American company Intuitive Surgical
Year of creation: 2000 (initial FDA approval)
most commonly for hysterectomies and prostate removals
increasingly for cardiac valve repair
Advantages: No trembling (neither surgeon nor
patient) Good visibility for laparoscopy very sharp picture of operational field
4
18.01.2017
S. Jeschke
Introduction
Classifications in robotics (selection) in general
highlow
Intelligence
(means: self-adaptivity; decision-making)
mobilestationary
Degree of mobility
high= robotdepending on human control
Low = robot fully autonomous
Cooperation level
5
18.01.2017
S. Jeschke
Introduction
Classifications in robotics for da Vinci
highlow
Intelligence
(means: self-adaptivity; decision-making)
mobilestationary
Degree of mobility
high= robotdepending on human control
Low = robot fully autonomous
Cooperation level
6
18.01.2017
S. Jeschke
Introduction
A more general approach to “medical robots”
rehabilitation / aftertreatmentprevention acute treatment
conservative
surgical
da Vinci(urology &
gynecology)
DLR MiroSurge(cardiac)
Renaissance by Mazor
(spine surgery)
7
18.01.2017
S. Jeschke
Introduction
An ever more general 4.0 approach to “medical robots”
Big Data Analytics
Hospital logistics
Patient care
Elderly care
Quantified self
Fluent transition to fitness/wellness
Humanoid robots
Robots in medical production (implants)
Robots as artificial limbs
Robots as trainers andphysiotherapists
Pet robots
8
18.01.2017
S. Jeschke
Outline
I. Introduction
da Vinci as a starting point
towards a more general approach to robots in medicine
II. Physical robots
in different application areas
in different shapes and design principles
III. Non-physical robots
in a nutshell
from Google Flu to Watson and beyond
IV. Summary and Outlook
the question of a creative artificial mind
4.0 trends in labor and employment
9
18.01.2017
S. Jeschke
Applications for robots in medicine
Intralogistics goes mobile: The Festo Logistics League
Competitions robocup:
2012: 0 points in World Cup
2013: 4th in World Cup
2014: Winner of the GermanOpen
2014: Winner of the World Cup
2015: Winner of the World Cup
2016: Winner of the World Cup
Critical factors for success: totally decentralized no ”hard coded components“ strong cooperation re-planning during tasks
Mobile transportation robots from flexible routing
!
Competencies: localization & navigation computer vision adaptive planning multi agent strategies sensors & hardware
10
18.01.2017
S. Jeschke
Applications for robots in medicine
Mobile intralogistics robots for medical environments
Robot by MLR, 2015, at Nye Akershus University Hospital Oslo:
automatic goods transport system
THORSTEN, IMA/ZLW & IfU @ RWTH: Based on the algorithms of the FESTO Logistics League
Hospital logistics
Car-O-Bot by Fraunhofer IPA, since about 1990
Spot, Boston Dynamics, background: military research (2015)
11
18.01.2017
S. Jeschke
Applications for robots in medicine
Human machine interaction and cooperative robotics
Robots are no longer locked in work-cells but cooperate with each other and/or with humans
Direct interaction – object transfer
Coping with visual occlusions of the object resulting from the body/ the hands of the human
Evaluating possible grasping points Offering good grasping options for the human co-worker
12
18.01.2017
S. Jeschke
Applications for robots in medicine
Into Service Robotics: The next step – the “Oscars”
Transform mobile robotic experiences into the field of service robotics
!
1. Investigating “new” human machine interfaces and interaction schemes Simple, intuitive Schematic eyes following you “natural eyes behavior”: randomly
looking around, showing interest by blinking, looking bored, …
!Performing service robot tasks Distribute brochures and serving drinks Path planning, room exploration, …
!
2. Investigating the “Uncanny Valley”: when features look almost, but not exactly, like natural beings, it causes a response of revulsion among the observers (Mori 1970)
3. Investigating diversity specific reactions (gender, age, culture) to artificial systems and in particular robots
13
18.01.2017
S. Jeschke
Shadow Dexterous Hand
Applications for robots in medicine
From embodiment … to humanoids
Zykov V., Mytilinaios E., Adams B., Lipson H. (2005) "Self-reproducing machines", Nature Vol. 435 No. 7038, pp. 163-164Bongard J., et al., Resilient Machines Through Continuous Self-Modeling, Science 314, 2006Lipson H. (2005) "Evolutionary Design and Evolutionary Robotics", Biomimetics, CRC Press (Bar Cohen, Ed.) pp. 129-155
Robonaut 2- NASA
The Bongard robot – learning through embodiment [Bongard, 2006; Lipson, 2007]
Embodiment theory:„intelligence needs a body“
The existence of a body (incl. sensors and actuators)are basic prerequisites to build experience and finally the development of intelligence.
Embodiment theory:„different bodies = different intelligences“
… leading to humanoids / humanoid components
Asimo Honda
KIT, Dillmann, SFB 588
14
18.01.2017
S. Jeschke
Applications for robots in medicine
Robots for teaching and training
Robot TOSY TOPIO
sport roboter“ – a „pal“ for leisure time but in general, the system can also be used
for rehabilitation training developed in Vietnam plays table tennis against human players
(and wins!) capable of learning, improving his style, but
also adapt to the opponent in question
Robot Sayah
Teaching roboter“ in classroom, so farfor languages
but in general, the system can also beused for all kind of instructions
developed in Japan Comprehensive mimics to illustrate
emotions like anger, surprise, happyness, fear, disgust, ....
Kids react strongly to the emotionsdisplayed
15
18.01.2017
S. Jeschke
Applications for robots in medicine
Robots as artificial replacement of limps
DEKA Bionic Arm “LUKE”
LUKE “Life Under Kinetic Evolution” Developed by DARPA Pentagon Agency) R&D program HAPTIX “seeks to create
a prosthetic hand system that moves and provides sensation like a natural hand”
Connection of human nerves to robotby surgery (Targeted musclereinnervation)
[IEEE Spectrum, Feb 2015, DARPA.MIL Dez 2016]
A Bionic Dance Prothesis
Developed by MIT BiomechatronicsGroup
Control system improves movement ofprosthetics by muscles
Allows „expressive activity modes“ similar to non-amputees
[PLOS ONE Aug 2015]
16
18.01.2017
S. Jeschke
Applications for robots in medicin
Support for nursing by robots
ROBEAR - RIKEN and Sumitomo Riko labs in Japan, to support hospital employees and people
in their private homes (2015)
!
The bearing robot ROBEAR
Lifts people from their beds, or into a wheelchair
Stationary so far, but in the future could be extended to a “logistists tool”, transferring patients
between rooms etc.
Decreases the physical stress for hospital employees
Supports an autonomous and self-determined life for people with handicaps
17
18.01.2017
S. Jeschke
Applications for robots in medicine
Pet robots
PARO therapeutic pet robot from Japan (movie 2011, designed since about 1993 by Takanori Shibata/AIST)
!
The therapeutic pet robot PARO
Mainly used for therapy for dementia patients
classified as a Class 2 medical device by U.S. regulators in fall 2009
responds to petting by moving its tail and opening and closing its eyes
actively seeks out eye contact, responds to touch, cuddles with people, remembers faces, …
?
Background - contact toanimals is known for itspositive impacts:
creation of meaning sozial catalyst relaxation motorical stimulation higher stress resistance stress reduction control of blood pressure
→ Enhances quality of lifeJustoCat, Robyn
Robotics AB/Sweden, Class 1 medical device
18
18.01.2017
S. Jeschke
Outline
I. Introduction
daVinci as a starting point
Towards a more general approach to robots in medicine
II. Physical robots
in different application areas
in different shapes and design principles
III. Non-physical robots
in a nutshell
from Google Flu to Watson and beyond
IV. Summary and Outlook
The question of a creative artificial mind
4.0 trends in labor and employment
19
18.01.2017
S. Jeschke
Applications for non-physical robots in medicine
Google Flu: predicting future (predicting the spread of diseases)
actual flu trend can be identified 7-10 days earlier by ‘Google Flu Trends’ than by official data of the Center for Disease Control (CDC)
[Helft 2008]
Analysing user behaviour
20
18.01.2017
S. Jeschke
Applications for non-physical robots in medicine
Predicting human behavior: transparent consumers
How Target figured out a teen girl was pregnant before her father did…
Unique Target IdEach interaction with retailer is assigned to that id
Customer profilesClustering customers into groups, for example to identify disruptions in life(e. g. weddings, job changes and pregnancy)
Andrew PoleStatistician working for Target
Pole identified about 25 products that allowed him to assign each customer a “pregnancy prediction” score and the estimated due date
Coupon campaign
Group of pregnant
customers
Customer
21
18.01.2017
S. Jeschke
Linguisticpreprocessing
Generation of possiblecandidates
Evaluation of candidates
Applications for non-physical robots in medicine
The new probabilistic engines
? Back to Watson: how is this guy running the (Jeopardy!) show??
DeepQA architecture Purely based on natural language processing (NLP) Approx. 100 different AI/linguistic methods come into play Without any specific semantic representation (“as-is”)
90 IBM-Power-750 servers For each: a 3.5 GHz POWER7
processor, with 8 cores, and 4 threads per core
In total: 2.880 POWER7 threads 16 terabytes of RAM
22
18.01.2017
S. Jeschke
Applications for non-physical robots in medicine
Probabilistic engines “down-to-earth”
! Watson: from playing Jeopardy! towards becoming some kind of a “medical doctor”…
today, only 20% of the medical knowledge is evidence based (basis of individualized medicine)
also, amount of medical information is doubling every 5 years: physicians can’t read all the journals
Data: all types, 1. structured data from electronic medical record
databases and 2. unstructured text from physician notes and
published literature
How can we deal with these challenges?
Goal of Watson: help physicians in diagnosing and treating patients by analyzing large data
acting as a huge preprocessor for all kind of medical information
potential to transform health care into individual medicine
currently tested by several clinics, e.g. Mayo, MD Anderson, Cleveland, and Sloan-Kettering
“IBM's Watson is better at diagnosing cancer than human doctors”
Example “p53”: Watson identified possible treatments for protein p53 deficiency linked to many cancers
Example “Google Flu” (another engine): already now, doctors integrate the results of GoogleFlu (spreading and direction of contagious illnesses) as it is much faster and more precise as the results of the best medical centers in the world)
23
18.01.2017
S. Jeschke
Outline
I. Introduction
da Vinci as a starting point
towards a more general approach to robots in medicine
II. Physical robots
in different application areas
in different shapes and design principles
III. Non-physical robots
in a nutshell
from Google Flu to Watson and beyond
IV. Summary and Outlook
the question of a creative artificial mind
4.0 trends in labor and employment
24
18.01.2017
S. Jeschke
Learning by doing – reinforcement learning
The next step: Using rewards to learn actions
?Remember Mario: What if the machine could learn how to solve a level? Why not use a some kind of intelligent trial-and-error?
Reinforcement learning (R-learning) is inspired by behaviorist psychology –maximizing the expected return by applying a sequence of actions at a current state.
Central part of cybernetics from its start (e.g., Minsky 1954)
[SethBling, 2015]
Neuroevolution of augmenting topologies (NEAT)
Genetic algorithms on top of neural networks
At each state the system decides what action to do
Actions are rewarded if Mario does not die in return
Level progress by evolving neural networks
[Stanley, 2002]
Now, Human factor is reduced to very general, formal
specifications of the neural network… However, human still influences the
underlying representation model
25
18.01.2017
S. Jeschke
The creative artificial mind
Where the Story Goes: AlphaGo
!
Go originated in China more than 2,500 years ago. Confucius wrote about it. As simple as the rules are, Go is a game of profound complexity. This complexity is what makes Go hard for computers to play, and an irresistible challenge to AI researchers. [adapted from Hassabis, 2016]
Bringing it all together!
The problem: 2.57×10210 possible positions – that is more than the number of atoms in the universe, and more than a googol times (10100) larger than chess.
Training set30 million moves recorded fromgames played by humans experts
Creating deep neural networks12 network layers with millions ofneuron-like connections
Predicting the human move(57% of time)
Dat
a-d
rive
n le
arn
ing
Re
info
rce
me
nt
lear
nin
g
Learning non-human strategiesAlphaGo designed by Google DeepMind, played against itself in thousands of games and evolved its neural networks; Monte Carlo tree search
! Achieving one of the grand challenges of AI
March 2016:Beating Lee Se-dol (World Champion)AlphaGo won 4 games to 1.(5 years before time)
[Has
sab
is, 2
01
6]
26
18.01.2017
S. Jeschke
!“Creativity is a phenomenon whereby something new … is formed. The created item may be intangible (such as an idea, a scientific theory, a musical composition or a joke) or a physical object (such as an invention, a literary work or a painting).” [adapted from Wikipedia, last visited 5/3/2016]
The creative artificial mind
Microsoft Visual Storytelling (SIS): machines becoming creative
Visual-Storytelling by Microsoftbased on deep neural networks(convolutional neural networks)
DII (descriptions for images in isolation): Traditional storytelling software
SIS (stories for images in sequence): new approach towards storytelling, including
Based on SIND – Sequential Image Narrative Dataset: 81,743 unique photos in 20,211 sequences, aligned to both descriptive (caption) and story language.
[Margaret Mitchell / Microsoft, 04/2016, together with colleagues from Facebook]
27
18.01.2017
S. Jeschke
!“Creativity is a phenomenon whereby something new … is formed. The created item may be intangible (such as an idea, a scientific theory, a musical composition or a joke) or a physical object (such as an invention, a literary work or a painting).” [adapted from Wikipedia, last visited 5/3/2016]
Van Gogh’s Starry Nightinterpreted by Google DeepDream
based on deep neural networks
“Do Androids Dream of Electric Sheep?”
(science fiction novel by American writer Philip K. Dick, published in 1968)
Computational creativity (artificial creativity) … is a multidisciplinary endeavor that is located at the intersection of the fields of artificial intelligence, cognitive psychology, philosophy, and the arts. [adapted from Wikipedia, last visited 5/3/2016]
„Can machines be creative?“ by Iamus, a computer clustercomposing classicalmusic by genetic algorithms, concert forTurings 100th birthday [youtube]
The creative artificial mind
Google DeepDream: machines becoming creative
28
18.01.2017
S. Jeschke
Changes to the job market
Industry 4.0 does not only change the “routine” jobs
The typical assumption…
… that job changes in 4.0 are mainly addressing blue collar jobs and/or routine jobs does not hold true.
White collar jobs
… are under massive change due to the enhancement in
AI, here the impact often hits “middle class jobs”
Social robots
… will become capable of taking over even complex
tasks with personal presence as in health or home care
From „blue collar – low qualified“ to „white collar – middle class“...
but probably, this is just a transition phenomenon
High qualified jobs
… as e.g. health professionals face already the taking over
through AI in certain fields by Watson, Google Flu, etc.
Decentralized platforms
… with automated consensus models (e.g. blockchain) take over complex administrative
tasks e.g. in judiciaries
Virtual and augmented environments
… allowing for new international players, even in tasks requiring humans
and presence
Autonomous systems
… as autonomous cars and more envanced production
technology will change the blue collar – low qualified as well
www.ima-zlw-ifu.rwth-aachen.de
Thank you!Univ.-Prof. Dr. rer. nat. Sabina JeschkeHead of Cybernetics Lab IMA/ZLW & [email protected]
Co-authored by:
Prof. Dr.-Ing. Tobias MeisenJunior Professor “Interoperability of Simulations”
Dipl.-Inform. Christian KohlscheinHead of Research group Cognitive Computing & eHealth
Dr. rer. nat. Katja SchneiderPersonal Assistance to the Heads of Institutes