Download Artificial-intelligence-augmented clinical medicine

Artificial-intelligence-augmented clinical medicine
Klaus-Peter Adlassnig
Section for Medical Expert and Knowledge-Based
Systems
Center for Medical Statistics, Informatics, and
Intelligent Systems
Medical University of Vienna
Spitalgasse 23, A-1090 Vienna
www.meduniwien.ac.at/kpa
Einführung in Medizinische Informatik, WS 2013/14, 30. Oktober 2013
Artificial Intelligence (AI)
• Definition 1: AI is a field of science and engineering concerned with the
computational understanding of what is commonly called intelligent behavior,
and with the creation of artifacts that exhibit such behavior.
from: Shapiro, S.C. (1992) Artificial Intelligence. In Shapiro, S.C. (ed.) Encyclopedia of Artificial Intelligence,
2nd ed., vol. 1, Wiley, New York, 54–57.
• Definition 2: AI is the science of artificial simulation of human thought
processes with computers.
from: Feigenbaum, E.A. & Feldman, J. (eds.) (1995) Computers & Thought. AAAI Press, Menlo Park, back cover.
Artificial Intelligence—applicable to clinical medicine
• It is the decomposition of an entire clinical thought process and its separate
artificial simulation—also of simple instances of “clinical thought”—that make
the task of AI in clinical medicine manageable.
• A functionally-driven science of AI that extends clinicians through computer
systems step by step can immediately be established.
⇓
artificial-intelligence-augmented clinical medicine
Computational intelligence in medical research
medical knowledge
definitional, causal, statistical, and heuristic knowledge
facts
molecular
biomedicine
consensus
evidence-based
medicine
biomolecular research
medical statistics
clustering and classification
data and knowledge mining
consensus conferences
generalization
medical
studies
Computational intelligence in patient care
human-to-human
patientphysician
encounter
AI-augmentation
decision-oriented
analysis and interpretation
of patient data
medical
knowledge modules
definitional, causal, statistical, and heuristic knowledge
medical knowledge
human-to-human
patientphysician
follow-up
Computers in clinical medicine—steps of natural progression
• step 1: patient administration
• admission, transfer, discharge, and billing
• step 2: documentation of patients’ medical data
• electronic health record: all media, distributed, life-long (partially fulfilled)
• step 3: patient and hospital analytics
• data warehouses, quality measures, reporting and research databases,
patient recruitment
… population-specific
• step 4: clinical decision support
• safety net, quality assurance, evidence-based
… patient-specific
Clinical medicine
medication history
symptomatic therapy
history data
symptoms
symptoms
signs
biosignals
images
genetic data
signs
laboratory
diagnosis
…
patient
laboratory
test results
test
results
differential
diagnosis
differential
therapy
findings
radiological
diagnosis
medical guidelines
examination
subspecialities
clinic
prognosis
patient
Clinical medicine
medication history
symptomatic therapy
history data
signs
…
…
biosignalsANNs
…
patient
MYCIN
laboratory
laboratory
test results
diagnosis
images SVMs
radiological
diagnosis
genetic data
personalized
medicine
examination
subspecialities
signs
test
results
…
QMR
DxPlain
CADIAG
symptoms
symptoms
+differential
diagnosis
differential
therapy
findings
+
medical guidelines
clinic
prognosis
patient
Clinical medicine: high complexity
•
•
•
•
sources of medical knowledge
‒
definitional
‒
causal
‒
statistical
‒
heuristic
layers of medical knowledge
‒
observational and measurement level
‒
interpretation, abstraction, aggregation, summation
‒
pathophysiological states
‒
diseases/diagnoses, therapies, prognoses, management decisions
imprecision, uncertainty, and incompleteness
‒
imprecision (=fuzziness) of medical concepts
*
due to the unsharpness of boundaries of linguistic concepts
‒
uncertainty of medical conclusions
*
due to the uncertainty of the occurrence and co-occurrence of imprecise medical concepts
‒
incompleteness of medical data and medical theory
*
due to only partially known data and partially known explanations for medical phenomena
“gigantic” amount of medical data and medical knowledge
‒
patient history, physical examination, laboratory test results, clinical findings
‒
symptom-disease relationships, disease-therapy relationships, …
‒
terminologies, ontologies: SNOMED CT, LOINC, UMLS, …
specialisation, teamwork, quality management, computer support
• studies in Colorado and Utah and in New York
(1997)
– errors in the delivery of health care leading to
the death of as many as 98,000 US citizens
annually
• causes of errors
– error or delay in diagnosis
– failure to employ indicated tests
errors
– use of outmoded tests or therapy
– failure to act on results of testing or
monitoring
– error in the performance of a test, procedure,
or operation
– error in administering the treatment
– error in the dose or method of using a drug
– avoidable delay in treatment or in responding
to an abnormal test
prevention
– inappropriate (not indicated) care
– failure of communication
– equipment failure
• prevention of errors
– we must systematically design safety into
processes of care
Medical information and knowledge-based systems
patient’s medical data
symptoms
signs
test results
clinical findings
biosignals
images
diagnoses
therapies
nursing data
•
•
•
standardization
telecommunication
chip cards
medical statistics
clustering & classification
data & knowledge mining
machine learning
induction
many
patients
•
•
•
general
knowledge
clinical decision support
medical expert systems
deduction
single
patient
general
knowledge
anatomy
biochemistry
physiology
pathophysiology
pathology
nosology
therapeutic knowledge
disease management
•
•
•
subjective experience
intuition
diagnosis
therapy
prognosis
management
knowledge-based
systems
information
systems
telemedicine
physician’s medical knowledge
integration
telemedicine
Clinical decision support and quality assurance (in general)
patients’ structured medical data
diagnostic support
• clinical alerts, reminders, calculations
• data interpretation, (tele)monitoring
• differential diagnostic consultation
– rare diseases, rare syndromes
– further or redundant investigations
– pathological signs accounted for
• consensus-criteria-based evaluation
– definitions
– classification criteria
prognostic prediction
• illness severity scores, prediction rules
• trend detection and visualization
therapy advice
• drug alerts, reminders, calculations
– indication, contraindications,
redundant medications, substitutions
– adverse drug events, interactions,
dosage calculations, consequent orders
• management of antimicrobial therapies, resistance
• (open-loop) control systems
patient management guidelines & quality assurance
• evidence-based reminders and processes
• computerized clinical guidelines, protocols, SOPs
• healthcare-associated infection surveillance
according to
Kensaku Kawamoto,
University of Utah, 2012:
“A Holy Grail of clinical
informatics is scalable,
interoperable clinical
decision support.”
What have we done?
Interpretation
of
hepatitis serology test results
test results
interpretation
ORBIS Experter: Hepatitis serology diagnostics
Interpretation
of
hepatitis serology test results
Differentialdiagnose
rheumatischer
Erkrankungen
Computergestützte
Entwöhnung vom
Respirator
Personalized clinical decision support
patient medical data
history
physical
signs
lab
tests
clinical
findings
+
genomic
data
present (personalized) CDS
future personalized CDS
unavoidable, more specific diagnostics, extends the realm of therapy
Solution at the Vienna General Hospital
hospital
Arden Syntax
development &
test environment
wards
HIS
knowledge
HIS
units
PDMS
clinical
laboratories
extended
documentation
& research
data
base
Arden Syntax server
knowledge
interfaces
departments
data &
knowledge
services center
LIS
genomic data
laboratories
LIS
data & concept
mining
results
data
Arden Syntax
rule engine
Arden Syntax and Health Level Seven (HL7)
• A standard language for writing situation-action rules that can trigger alerts
based on abnormal clinical events detected by a clinical information system.
• Each module, referred to as a Medical Logic Module (MLM), contains sufficient
knowledge to make a single decision.
 extended by packages of MLMs for complex clinical decision support
• The Health Level Seven Arden Syntax for Medical Logic Systems, Version 2.9—
including fuzzy methodologies—was approved by the American National
Standards Institute (ANSI) and by Health Level Seven International (HL7) on
14 March 2013.
 continuous development since 1989
General MLM Layout
Maintenance Category
Library Category
Knowledge Category
Resources Category
Identify an MLM
Data Types
Operators
Basic Operators
Curly Braces
List Operators
Logical Operators
Comparison Operators
String Operators
Arithmetic Operators
Other Operators
Control Statements
Call/Write Statements and Trigger
Sample MLM (excerpt)
Arden Syntax, Arden Syntax server, and health care information
integration
systems
HIS, MIS, PDMS, LIS,
medical practice SW,
web-based EHR,
telemedicine
applications,
health portals,
…
service-oriented
*data & knowledge services center
operational:
- harmonized input data
- Arden Syntax MLMs
- collected reasoning data
exploratory:
- rule learning/tuning
- data and concept mining
*
web-based
functionality
reminders and alerts,
monitoring,
surveillance,
diagnostic and
therapeutic decision
support,
…
Arden Syntax server and software components
Arden Syntax
development & test
environment
knowledge
Arden Syntax server 1)
results
reporting
tools
knowledge
administration
interfaces 2)
health care
information system
• Arden Syntax integrated
development and test
environment (IDE)
including
‒ Medical logic module
(MLM) editor and
authoring tool
‒ Arden Syntax compiler
(syntax versions 2.1,
2.5, 2.6, 2.7, 2.8, and
2.9)
‒ Arden Syntax engine
‒ MLM test environment
‒ MLM export component
• command-line
Arden Syntax compiler
knowledge
data & knowledge
services center
results
data
data
1)
2)
integrated, local, or remote
local and web services, web frontend
Arden Syntax
rule engine
• web-services-based
Arden Syntax server
including
‒ Arden Syntax engine
‒ MLM manager
‒ XML-protocol-based
interfaces, e.g., SOAP,
REST, and HL7
‒ a project-specific data
and knowledge services
center may be hosted
• Java libraries
‒ Arden Syntax compiler
‒ Arden Syntax engine
Fuzzy Arden Syntax: Modelling uncertainty in medicine
• linguistic uncertainty
‒ due to the unsharpness (fuzziness) of boundaries of linguistic concepts;
gradual transition from one concept to another
‒ modeled by fuzzy sets, e.g., fever, increased glucose level
• propositional uncertainty
‒ due to the uncertainty (or incompleteness) of medical conclusions;
includes definitional and causal, statistical and subjective relationships
‒ modeled by truth values between zero and one, e.g., usually, almost
confirming
Crisp sets vs. fuzzy sets
yes/no decision
χY
young
U = [0, 120]
Y ⊆ U with Y = {(χY (x)/ x)x ∈ U}
χ Y: U → {0, 1}
1
0
0
age
 1
x > threshold
x ≤ threshold
∀x∈U
gradual transition
µY
young
1
0
threshold

χ Y (x) =  0
0
threshold
age
U = [0, 120]
Y ⊆ U with Y = {(µY (x)/ x)x ∈ U}
µY: U → [0, 1]
1
1 + (0.04
x > threshold
x)2
µY (x) =
∀x∈U
1
x ≤ threshold

Crisp sets vs. fuzzy sets
χY
young
“arbitrary” yes/no decisions
1
0
0
µY
age
young
1
0
threshold
0
threshold
• cause of unfruitful
discussions
• often simply wrong
“intuitive” gradual transitions
age
Degree of compatibility [= degree of membership]
µA (x)
1.00
highly
decreased
↓↓
decreased
↓
normal
increased
⊥
↑
highly
increased
↑↑
µ↑ (x) = 0.82
0.50
µ⊥ (x) = 0.18
µ↓↓ (x) = 0.00
µ↓ (x) = 0.00
µ↑↑ (x) = 0.00
0.00
50
100
glucose level in serum of 130 mg/dl
150
200
x [mg/dl]
Integration into i.s.h.med
at the
Vienna General Hospital
SOP checking
in melanoma patients
receiving chemotherapy
Towards a science of clinical medicine
patient’s medical data
and
healthcare processes
for
human processing
observations
e.g., temperature chart
skin color (jaundice, livid, …)
…
≠
patient’s medical data
and
healthcare processes
for
machine processing
measurements
e.g., CRP
color measurement
…
“Measure what is measurable, and make measurable what is not so.”
Galileo Galilei
1564–1642
Crucial point in clinical medicine:
“Digitize what is digitizable, and make digitizable what is not so.”
Klaus-Peter Adlassnig
The medical world becomes flat
after Thomas L. Friedman
The World is Flat.
Penguin Books, 2006.
• in a local world
‒ decision support will be part of clinical information systems
• in a global world
‒ any activity—where we can digitize and decompose the value chain*,
and move the work around—will get moved around
* patient value chain: patient examination, diagnosis, therapy, prognosis, health care
decisions, patient care
Future of artificial-intelligence-augmented clinical medicine
medical data
collection,
storage,
&
distribution
0
today
tomorrow
clinical
decision
support
personalized
medicine
predictable
future
implants,
prostheses,
robotics
1
1
3
4
2
4
HIS 1.0
HIS 2.0
1
Closing remark: formalism vs. reality
Pure mathematics is much easier to understand, much simpler, than the
messy real world!
Gregory Chaitin (2005) Meta Math!: The Quest for Omega,
Pantheon Books, New York.
Clinical informatics deals with the “messy real patient”.