Transcript CSE5810 - University of Connecticut

Artificial Intelligence & Clinical Decision Support.
Including fuzzy logic, neural nets, and genetic algorithms
CSE
5810
Kevin Lopez
Computer Science & Engineering Department
The University of Connecticut
371 Fairfield Road,
Storrs, CT 06269-2155
[email protected]
Lopez-1
What is Clinical Decision Support?

CSE
5810


Clinical Decision Support is:
 Knowledge provided to clinicians
 From Multiple Sources/Contexts, processed and
returned in a form that will assist a care giver.
 Involves processing via various artificial
intelligence and machine learning technologies
CDS is Multi-Disciplinary
 Computing (Information Processing, Data
Analysis)
 Social Science (User Interactions)
Clinical Decision is applicable to many domains:
 Can be used in any type of medicine, including
domains with weak domain theory.
 Its underlying systems (AI) is used for any field
Lopez-2
What is Clinical Decision Support?

CSE
5810


Key people(s) affected:
 Patients
 Physicians, clinicians, care givers
 Hospitals/medical centers
Standards:
 Arden Syntax (Syntax)
 GELLO (Common Expression Language)
 Infobutton (Context-aware Knowledge Retrieval)
Techniques:
 These are still being worked on and researched.
 No set technique
Lopez-3
What is Clinical Decision Support?

CSE
5810
A combination of
different knowledge's.
 Knowledgebase
(Textbook, etc)
 Clinicians
Knowledge/experien
ce
 Gained Experience
from learning, and
individual patients
Clinical Decision
Support System
Knowledgebase
Clinician
Gained Experience
Lopez-4
Why use a CDSS?

CSE
5810
We use a CDSS because of:
 It provides better quality of care
 Can provide the clinician with a second opinion
 Can guide a novice clinician to a solution,
diagnosis, or treatment.
 Can help reduce the number of errors
 It can help with the speed and quality of diagnosis
 It improves customer/patient satisfaction
 Can be interactive (with the clinician) to get the
best results.
 Can be nearly autonomous, some systems are
personal and can give a diagnosis.
Lopez-5
Functions of a CDSS

CSE
5810
A CDSS generally works by:
 Taking in some data, normally it is some patient
data
 This data can be measurements, clinician data, or
knowledgebase data.
 The data then must be extrapolated and the most
relevant parts used for processing.
 The data is then processed with the method of
choice (ANN, CBR, Fuzzy etc.) and may require
clinician input as well.
 The data is then post processed and outputted in a
variety of fashions (can be numerical, binary, or
even text).
Lopez-6
Designing a CDSS

CSE
5810
Main problems these systems must solve
 Structured
 These problems are routine and repetitive
 Solutions exist, and are standard and predefined

Unstructured
 Complex and fuzzy
 Lack Clear and straightforward solutions

Semi-structured
 This is a combination of the two previous catagories.
Lopez-7
Artificial intelligence's role in clinical decision support

CSE
5810


Two types of CDSS
 Work with Knowledgebase
 Work with Non-Knowledgebase
Knowledge based CDSS:
 Use knowledge from sources such as textbooks,
and other resources.
 They have rules similar to if-then statements.
Components of a knowledge based CDSS:
 Knowledgebase: Some source where they get their
knowledge
 Inference engine: takes data and applies the rules
from the knowledgebase
 Communication: Allows system to communicate
with user and user input.
Lopez-14
Hybrid Systems

CSE
5810
Hybrid systems Knowledge and Non-Knowledge
based system
 These systems produce high quality results from
the merge of the two different systems.
 They have an already established knowledge base
but they also must learn from past experiences or
from test results.
 These systems often Produce results that are better
than these systems individually.
 These systems can be a combination of many of
the different technologies that each system has.
Lopez-16
Artificial Neural Networks

CSE
5810
Similar to real neural
networks
 Take in data and
pass them through
the network to the
other neurons to
get an output.
 Many times used
for pattern
recognition
 Several different
algorithms can be
used for threshold
Lopez-18
Case-Based Reasoning

CSE
5810

Case-based reasoning is:
 A process of solving new problems based off of
old problems.
 Similar to how humans think and solve problems.
 Can take new solutions that have been solved and
add them to the database of solutions for future
reference.
There are Four Steps (R’s) to case based reasoning:
 Retrieve: where the system retrieves the
knowledge
 Reuse: takes old experience and maps it to new
problem
 Revise: revise the solution
 Retain: put new solution into the system database
Lopez-22
Case-Based Reasoning

The four R’s for Case based reasoning
CSE
5810
Lopez-27
Fuzzy Techniques

CSE
5810

Fuzzy Logic is:
 Degrees of truth, 0 and 1 are extremes.
 Some types of data do not have what we consider a
full truth or false.
An example of Fuzzy Logic
 An example of this is natural language processing.
 This is where truths are aggregated from partial
truths.
 This is to derive meaning from humans such as
notes a doctor put in or some other source of
natural language.
Lopez-28
Genetic Algorithms

CSE
5810


Based off of a simplified evolutionary process used to
arrive at an optimal solution.
It works in the following way:
 Children are made and try to solve the problem
 The top few children then are used to generate new
children
 This process continues until an optimal (or very
close to optimal) solution is found.
In CDSS:
 The selected algorithms evaluate the solution
 Of these solutions the best are chosen and they try
to evaluate the problem again until the solution is
found.
Lopez-30
Feature Selection

CSE
5810

Feature Selection is:
 Selecting features or attributes from a set of data
 Useful for taking out certain data that is not needed
during processing
 Similar to how we process data, we do not need to
know all of the data but we extract key items from
the data.
 Data may have redundant features that provide no
more information as the features previously
selected.
Feature Selection is used in getting the data that is
required.
 Allows for less and unnecessary processing.
Lopez-31
Personal Medicine

CSE
5810

There are several apps that claim to assist with
diagnosis.
In particular several skin cancer apps have surfaced.
 None of which are free
 Some of which incorporate sending the images to
a clinician for further diagnosis.
 Some of the apps have the ability to use the
camera to view the skin and take a picture
 With this picture the program checks for
symptoms, or “ugly duckling moles”
 Apps are still improving to give more quality care
Lopez-35
Personal Medicine
CSE
5810
Lopez-36
Effectiveness of CDSS

CSE
5810
How effective are these systems
 CDSS’s are becoming more and more effective
and accurate at diagnosing diseases.
 Many times these systems improve the outcome of
both treatments and diagnosis of patients
 Many times these systems are integrated into the
clinicians workflow to provide superior
satisfaction to both the patient and the clinician.
 These systems give the clinician a recommendation
not just an assessment, so that the clinician can
actually follow through.
 These systems many times outperform their
clinician counterparts in diagnosing a patient.
Lopez-39
Key Technical Problems

CSE
5810
Some of the problems that are seen with CDSS
 Many different types of artificial intelligence that
serve many different purposes
 No one generic algorithm that can handle all of the
data
 Natural language can be very difficult to extract
data from
 Some domains have weak domain theory
 Many of the systems need time to train and much
of the training is computationally expensive
 Data preferred to be shortened (feature selection)
in order to take less time processing.
Lopez-42
Key People Problems

CSE
5810
There are problems that exist where the user may
experience either due to lack of experience or
familiarity.
 Ease of use: The system must be easy to use, and
work right out of the box.
 There should be minimal configuration if any done
by the clinician.
 The interface has to be user friendly. Many times
users of these systems have very little computer
knowledge.
 The user should not have to be trained on this
system.
 Data input: the data must be entered correctly (ie.
switching systolic and diastolic).
Lopez-43
Conclusion

CSE
5810

These Systems Show:
 Improvement in patient outcome
 Higher Patient satisfaction
 Guidance for inexperienced practitioners
 Guidance for individuals
These systems cannot:
 Replace a doctor/care giver
 Are limited in how many different diseases each
one can do
 Be 100% accurate/fool proof
Lopez-44
References

CSE
5810













Application of Artificial Intelligence for Clinical Decision Making and Reasoning (Abdalla
S.A.Mohamed)
Efficient Clinical Decision Making by Learning from Missing Clinical Data (Farooq, Yang, Hussain,
Huang, MacRae, Eckl, Slack)
Developing Decision Support for Dialysis Treatment of Chronic Kidney Failure the researchers
explore and describe what goes into developing a CDS system for dialysis treatment.
Hybrid Case-Based System in Clinical Diagnosis and Treatment.
A Model to Predict Limb Salvage in Severe Combat-related Open Calcaneus Fractures
Clinical Decision support system for fetal Delivery using Artificial Neural Networks the team are using
ANN’s to assist doctors with decisions at critical times of fetal deliveries.
Implementing Decision Tree Fuzzy Rules in Clinical Decision Support System after Comparing with
Fuzzy based and Neural Network based systems
Case Studies on the Clinical Applications using Case-Based Reasoning
Improving clinical practice using clinical decision support systems: a systematic review of trials to identify
features critical to success (Kensaku Kawamoto, Caitlin A Houlihan, E Andrew Balas, David F Lobach)
Effects of Computerized Clinical Decision Support Systems on Practitioner Performance and Patient
Outcomes
E-Health towards Ecumenical Framework for Personalized Medicine via Decision Support System
Standards in Clinical Decision Support: Activities in Health Level Seven And Beyond
(https://www.dchi.duke.edu/conferences/posters-presentations/amia/2011-amia/KawamotoStandardsInClinicalDecisionSupport_slides.pdf)
Kai Goebel from Rensselaer Polytechnic Institute (http://www.cs.rpi.edu/courses/fall01/softcomputing/pdf/cbr1to3.pdf)
HealthIT (http://www.healthit.gov/policy-researchers-implementers/clinical-decision-support-cds)
Lopez-45