Integrative clinical decision support systems: foundations and implementations

Szymon Wilk

Uzyskaj dostęp

Zakup książkę

ISBN/ISSN:

978-83-7143-969-8

DOI:

Wydawnictwo:

Wydawnictwo Politechniki Poznańskiej

Rok wydania:

2010

Liczba stron:

139

XML:ISBN/ISSN:

ONIX MARC21

Bibliografia:

Wilk, Szymon. Integrative clinical decision support systems: foundations and implementations. Red. . Poznań: Wydawnictwo Politechniki Poznańskiej, 2010, 139 s. ISBN 978-83-7143-969-8

Bibliografia refWorks:

RT Book, Whole

SR Electronic(1)

A1 Wilk, S.

T1 Integrative clinical decision support systems: foundations and implementations

PB Wydawnictwo Politechniki Poznańskiej

PP Poznań

YR 2010

SN 978-83-7143-969-8

Bibliografia BibTex:

@Book{ 108329, author = "Wilk, Szymon", editor = "", title = "Integrative clinical decision support systems: foundations and implementations", publisher = "Wydawnictwo Politechniki Poznańskiej", year = "2010", address = "Poznań", isbn = "978-83-7143-969-8" }

Bibliografia endNote:

TY - BOOK

AU - Wilk, Szymon

ED -

TI - Integrative clinical decision support systems: foundations and implementations

PB - Wydawnictwo Politechniki Poznańskiej

CY - Poznań

PY - 2010

SN - 978-83-7143-969-8

ER -

Netografia (standard APA):

Wilk, Szymon. Integrative clinical decision support systems: foundations and implementations [baza danych online] Poznań: Wydawnictwo Politechniki Poznańskiej, 2010 [dostęp: 27 04 2024]. Dostęp w Ibuk Libra: https://libra.ibuk.pl/reader/integrative-clinical-decision-support-systems-foundations-and-szymon-wilk-108329.

knowledge management, clinical decision support system, expert system, software, artificial inteligence in medicine, knowledge system

The book is concerned with theoretical and methodological foundations of clinical decision support systems (CDSSs), i.e., systems designed to help healthcare professionals make clinical decisions. It is focused on patient-specific CDSSs tha...

Więcej

ISBN/ISSN:

978-83-7143-969-8

DOI:

Wydawnictwo:

Wydawnictwo Politechniki Poznańskiej

Rok wydania:

2010

Liczba stron:

139

XML:ISBN/ISSN:

ONIX MARC21

Acronyms and abbreviations6
Abstract8
1. Scope and objectives10
1.1. Scope10
1.2. Motivations11
1.3. Objectives13
1.4. Book organization15
1.5. Acknowledgments15
2. An architecture for an ICDSS and its theoretical foundations16
2.1. Introduction16
2.2. Conceptual model of decision making implying the MET3 architecture17
2.3. Related research on patient-specific CDSSs18
2.4. Methods for developing the MET3 architecture20
2.5. The MET3 architecture22
2.5.1. Requirements22
2.5.2. Goal model24
2.5.3. Domain ontology25
2.5.4. Agent model27
2.6. Discussion29
3. Methodologies for constructing diagnostic decision models34
3.1. Introduction34
3.2. Rough set theory36
3.2.1. Handling inconsistent and incomplete objects36
3.2.2. Identifying important condition attributes and assessing their information value39
3.2.3. Discovering rules in decision tables41
3.2.4. Using decision rules to classify new objects46
3.3. Proposed methods for induction of robust decision rules48
3.3.1. Modifications of existing rule induction algorithms48
3.3.2. Combination of LEM2* and EXPLORE* to handle imbalanced data51
3.4. Experimental evaluation of the proposed methods on clinical data53
3.5. Discussion56
4. A framework for indexing and retrieval of context-specific evidence58
4.1. Introduction58
4.2. Related research60
4.3. The Cochrane Library62
4.4. Proposed methods63
4.4.1. Design of the indexing and retrieval framework63
4.4.2. Phase 1: identification of core index concepts66
4.4.3. Phase 2: indexing of evidence-based documents67
4.4.4. Phase 3: retrieval of evidence-based documents68
4.5. Discussion70
5. Principles of designing interactions for an ICDSS74
5.1. Introduction74
5.2. Related research on interactions in CDSSs75
5.3. Methodological foundations of interaction design76
5.3.1. Framework of interactions76
5.3.2. Scenario-based design methodology77
5.3.3. Object-Action-Interface model77
5.4. Proposed principles for interactions in ICDSSs78
5.3.4. Eight Golden Rules of Interface Design78
5.5. Discussion83
6. Practical applications of the proposed methodologies86
6.1. Introduction86
6.2. Implementation of the MET3 architecture86
6.3. Generic process of constructing diagnostic decision models88
6.4. Construction of the diagnostic model for pediatric scrotal pain90
6.4.1. Description of the clinical problem90
6.4.2. Collected retrospective data91
6.4.3. Development of possible decision models92
6.4.4. Evaluation of selected decision models94
6.5. Construction of the diagnostic model for pediatric asthma exacerbations95
6.5.1. Description of the clinical problem95
6.5.2. Collected retrospective data96
6.5.3. Development of possible decision models99
6.5.4. Evaluation of selected decision models100
6.6. Implementation of the evidence indexing and retrieval framework for pediatric asthma exacerbations101
6.6.1. Identification of core index concepts101
6.6.2. Indexing evidence-based documents103
6.6.3. Retrieving evidence-based documents105
6.6.4. Evaluation of the automatic indexing106
6.7. Discussion108
7. Practical evaluation of a CDSS110
7.1. Introduction110
7.2. Methods112
7.2.1. Trial design and implementation112
7.2.2. Patient enrollment and follow-up114
7.2.3. Outcome measures and analysis115
7.3. Results116
7.3.1. Enrolled patients116
7.3.2. Accuracies of MET2-AP and physicians117
7.3.3. Inter-observer agreement117
7.4. Discussion119
8. Conclusions122
8.1. Results and contributions122
8.2. Future research directions123
References124
Streszczenie138

Scopeandobjectives

sionmaking[26,201].However,therearemanyfactorscomplicatingeffective

retrievalofevidencefromonlinerepositoriesofmedicalliterature.Theonline

medicalliteratureisvastanditsinvestigationcanrequiresignificanttimeand

effortonbehalfofthephysician.Otherfactorsincludeimpreciseindexing

schemaswhichfailtoadequatelydescribemedicalcontent,andsimplisticand

ineffectivequerymechanisms.Thus,thereisaneedforanewmethodologyfor

evidenceindexingandretrievalthatwilladdresstheseshortcomingsbyallowing

forautomaticprovision(withoutrequiringthephysiciantospecifyaquery)of

availableevidencerelevantinthecurrentcontextofapatient-physicianencoun-

ter.SuchmethodologycanbeemployedbyICDSSstoenhancerecommenda-

tionsgeneratedbydecisionmodelsorsuggestionselaboratedbyphysicianswith

relatedevidence,resultinginmoreinformeddecisions.

1.3.Objectives

Themajorobjectivesofthiscontributioncorrespondtothethreegroupsof

motivationspresentedinSection1.2andarethefollowing:

Toproposeanewconceptualsystemarchitecturethatcanbeusedasablu-

eprintfordevelopingICDSSs.Thisdesignwillallowconstructingsystems

thatsupportphysiciansincompletingalltasksinthedecisionmaking

processandareavailableatthepointofcare.

Toproposedataminingtechniquestoconstructdecisionmodelsfromclin-

icaldata,takingintoaccounttheundesirablecharacteristicsofsuchdata.

Theroughsetapproachallowsadequatehandlingofdataimperfections,

sinceitisabletodealwithincompleteandinconsistentdata,andto

representdiscoveredknowledgeinacomprehensibleformatofrobustde-

cisionrules.Createddecisionmodelsaremeanttobeembeddedin

ICDSSstosupportcliniciansindiagnostictasks.

Toproposeanewframeworkforindexingandretrievalofevidencethatis

relevantinthecurrentcontextofapatient-physicianencounter.Thisme-

thodwillsupplyevidencethatsupportsdiagnosisandtreatment,suggested

eitherbyaphysicianorbyasystem,andmatchesthecharacteristicof

apatient.Implementationoftheproposedmethodframeworkisplannedto

beembeddedinICDSSsforenhancedsupportofthetreatmentplanning

task.

Themajorobjectiveslistedabovecanbefurtherdecomposedintospecific

objectivesofscientificandpracticalnature.Theformerareconcernedwithpro-

posingnewtheoreticalsolutions(architectures,algorithmsandframeworks),and

thelatterfocusonpracticalimplementationandverificationofthetheoretical

proposals.Thespecificobjectivesofthisworkareasfollows:

Brak wyników

Integrative clinical decision support systems: foundations and implementations

Treść książki

Znajdź bibliotekę blisko siebie, i uzyskaj dostęp do ebooka w systemie IBUK Libra