Computer-aided diagnosis with potential application to rapid detection of disease outbreaks.
Identifieur interne : 001F24 ( PubMed/Corpus ); précédent : 001F23; suivant : 001F25Computer-aided diagnosis with potential application to rapid detection of disease outbreaks.
Auteurs : Tom Burr ; Frederick Koster ; Rick Picard ; Dave Forslund ; Doug Wokoun ; Ed Joyce ; Judith Brillman ; Phil Froman ; Jack LeeSource :
- Statistics in medicine [ 0277-6715 ] ; 2007.
English descriptors
- KwdEn :
- MESH :
- diagnosis : Anthrax, Hantavirus Infections.
- methods : Diagnosis, Computer-Assisted.
- Algorithms, Bioterrorism, Computer Simulation, Disease Outbreaks, Humans, Sensitivity and Specificity.
Abstract
Our objectives are to quickly interpret symptoms of emergency patients to identify likely syndromes and to improve population-wide disease outbreak detection. We constructed a database of 248 syndromes, each syndrome having an estimated probability of producing any of 85 symptoms, with some two-way, three-way, and five-way probabilities reflecting correlations among symptoms. Using these multi-way probabilities in conjunction with an iterative proportional fitting algorithm allows estimation of full conditional probabilities. Combining these conditional probabilities with misdiagnosis error rates and incidence rates via Bayes theorem, the probability of each syndrome is estimated. We tested a prototype of computer-aided differential diagnosis (CADDY) on simulated data and on more than 100 real cases, including West Nile Virus, Q fever, SARS, anthrax, plague, tularaemia and toxic shock cases. We conclude that: (1) it is important to determine whether the unrecorded positive status of a symptom means that the status is negative or that the status is unknown; (2) inclusion of misdiagnosis error rates produces more realistic results; (3) the naive Bayes classifier, which assumes all symptoms behave independently, is slightly outperformed by CADDY, which includes available multi-symptom information on correlations; as more information regarding symptom correlations becomes available, the advantage of CADDY over the naive Bayes classifier should increase; (4) overlooking low-probability, high-consequence events is less likely if the standard output summary is augmented with a list of rare syndromes that are consistent with observed symptoms, and (5) accumulating patient-level probabilities across a larger population can aid in biosurveillance for disease outbreaks.
DOI: 10.1002/sim.2798
PubMed: 17225213
Links to Exploration step
pubmed:17225213Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">Computer-aided diagnosis with potential application to rapid detection of disease outbreaks.</title>
<author><name sortKey="Burr, Tom" sort="Burr, Tom" uniqKey="Burr T" first="Tom" last="Burr">Tom Burr</name>
<affiliation><nlm:affiliation>Los Alamos National Laboratory, Mail Stop F600, Los Alamos, NM 87545, USA. tburr@lanl.gov</nlm:affiliation>
</affiliation>
</author>
<author><name sortKey="Koster, Frederick" sort="Koster, Frederick" uniqKey="Koster F" first="Frederick" last="Koster">Frederick Koster</name>
</author>
<author><name sortKey="Picard, Rick" sort="Picard, Rick" uniqKey="Picard R" first="Rick" last="Picard">Rick Picard</name>
</author>
<author><name sortKey="Forslund, Dave" sort="Forslund, Dave" uniqKey="Forslund D" first="Dave" last="Forslund">Dave Forslund</name>
</author>
<author><name sortKey="Wokoun, Doug" sort="Wokoun, Doug" uniqKey="Wokoun D" first="Doug" last="Wokoun">Doug Wokoun</name>
</author>
<author><name sortKey="Joyce, Ed" sort="Joyce, Ed" uniqKey="Joyce E" first="Ed" last="Joyce">Ed Joyce</name>
</author>
<author><name sortKey="Brillman, Judith" sort="Brillman, Judith" uniqKey="Brillman J" first="Judith" last="Brillman">Judith Brillman</name>
</author>
<author><name sortKey="Froman, Phil" sort="Froman, Phil" uniqKey="Froman P" first="Phil" last="Froman">Phil Froman</name>
</author>
<author><name sortKey="Lee, Jack" sort="Lee, Jack" uniqKey="Lee J" first="Jack" last="Lee">Jack Lee</name>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">PubMed</idno>
<date when="2007">2007</date>
<idno type="RBID">pubmed:17225213</idno>
<idno type="pmid">17225213</idno>
<idno type="doi">10.1002/sim.2798</idno>
<idno type="wicri:Area/PubMed/Corpus">001F24</idno>
<idno type="wicri:explorRef" wicri:stream="PubMed" wicri:step="Corpus" wicri:corpus="PubMed">001F24</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en">Computer-aided diagnosis with potential application to rapid detection of disease outbreaks.</title>
<author><name sortKey="Burr, Tom" sort="Burr, Tom" uniqKey="Burr T" first="Tom" last="Burr">Tom Burr</name>
<affiliation><nlm:affiliation>Los Alamos National Laboratory, Mail Stop F600, Los Alamos, NM 87545, USA. tburr@lanl.gov</nlm:affiliation>
</affiliation>
</author>
<author><name sortKey="Koster, Frederick" sort="Koster, Frederick" uniqKey="Koster F" first="Frederick" last="Koster">Frederick Koster</name>
</author>
<author><name sortKey="Picard, Rick" sort="Picard, Rick" uniqKey="Picard R" first="Rick" last="Picard">Rick Picard</name>
</author>
<author><name sortKey="Forslund, Dave" sort="Forslund, Dave" uniqKey="Forslund D" first="Dave" last="Forslund">Dave Forslund</name>
</author>
<author><name sortKey="Wokoun, Doug" sort="Wokoun, Doug" uniqKey="Wokoun D" first="Doug" last="Wokoun">Doug Wokoun</name>
</author>
<author><name sortKey="Joyce, Ed" sort="Joyce, Ed" uniqKey="Joyce E" first="Ed" last="Joyce">Ed Joyce</name>
</author>
<author><name sortKey="Brillman, Judith" sort="Brillman, Judith" uniqKey="Brillman J" first="Judith" last="Brillman">Judith Brillman</name>
</author>
<author><name sortKey="Froman, Phil" sort="Froman, Phil" uniqKey="Froman P" first="Phil" last="Froman">Phil Froman</name>
</author>
<author><name sortKey="Lee, Jack" sort="Lee, Jack" uniqKey="Lee J" first="Jack" last="Lee">Jack Lee</name>
</author>
</analytic>
<series><title level="j">Statistics in medicine</title>
<idno type="ISSN">0277-6715</idno>
<imprint><date when="2007" type="published">2007</date>
</imprint>
</series>
</biblStruct>
</sourceDesc>
</fileDesc>
<profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Algorithms</term>
<term>Anthrax (diagnosis)</term>
<term>Bioterrorism</term>
<term>Computer Simulation</term>
<term>Diagnosis, Computer-Assisted (methods)</term>
<term>Disease Outbreaks</term>
<term>Hantavirus Infections (diagnosis)</term>
<term>Humans</term>
<term>Sensitivity and Specificity</term>
</keywords>
<keywords scheme="MESH" qualifier="diagnosis" xml:lang="en"><term>Anthrax</term>
<term>Hantavirus Infections</term>
</keywords>
<keywords scheme="MESH" qualifier="methods" xml:lang="en"><term>Diagnosis, Computer-Assisted</term>
</keywords>
<keywords scheme="MESH" xml:lang="en"><term>Algorithms</term>
<term>Bioterrorism</term>
<term>Computer Simulation</term>
<term>Disease Outbreaks</term>
<term>Humans</term>
<term>Sensitivity and Specificity</term>
</keywords>
</textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">Our objectives are to quickly interpret symptoms of emergency patients to identify likely syndromes and to improve population-wide disease outbreak detection. We constructed a database of 248 syndromes, each syndrome having an estimated probability of producing any of 85 symptoms, with some two-way, three-way, and five-way probabilities reflecting correlations among symptoms. Using these multi-way probabilities in conjunction with an iterative proportional fitting algorithm allows estimation of full conditional probabilities. Combining these conditional probabilities with misdiagnosis error rates and incidence rates via Bayes theorem, the probability of each syndrome is estimated. We tested a prototype of computer-aided differential diagnosis (CADDY) on simulated data and on more than 100 real cases, including West Nile Virus, Q fever, SARS, anthrax, plague, tularaemia and toxic shock cases. We conclude that: (1) it is important to determine whether the unrecorded positive status of a symptom means that the status is negative or that the status is unknown; (2) inclusion of misdiagnosis error rates produces more realistic results; (3) the naive Bayes classifier, which assumes all symptoms behave independently, is slightly outperformed by CADDY, which includes available multi-symptom information on correlations; as more information regarding symptom correlations becomes available, the advantage of CADDY over the naive Bayes classifier should increase; (4) overlooking low-probability, high-consequence events is less likely if the standard output summary is augmented with a list of rare syndromes that are consistent with observed symptoms, and (5) accumulating patient-level probabilities across a larger population can aid in biosurveillance for disease outbreaks.</div>
</front>
</TEI>
<pubmed><MedlineCitation Status="MEDLINE" Owner="NLM"><PMID Version="1">17225213</PMID>
<DateCompleted><Year>2007</Year>
<Month>05</Month>
<Day>29</Day>
</DateCompleted>
<DateRevised><Year>2009</Year>
<Month>11</Month>
<Day>03</Day>
</DateRevised>
<Article PubModel="Print"><Journal><ISSN IssnType="Print">0277-6715</ISSN>
<JournalIssue CitedMedium="Print"><Volume>26</Volume>
<Issue>8</Issue>
<PubDate><Year>2007</Year>
<Month>Apr</Month>
<Day>15</Day>
</PubDate>
</JournalIssue>
<Title>Statistics in medicine</Title>
<ISOAbbreviation>Stat Med</ISOAbbreviation>
</Journal>
<ArticleTitle>Computer-aided diagnosis with potential application to rapid detection of disease outbreaks.</ArticleTitle>
<Pagination><MedlinePgn>1857-74</MedlinePgn>
</Pagination>
<Abstract><AbstractText>Our objectives are to quickly interpret symptoms of emergency patients to identify likely syndromes and to improve population-wide disease outbreak detection. We constructed a database of 248 syndromes, each syndrome having an estimated probability of producing any of 85 symptoms, with some two-way, three-way, and five-way probabilities reflecting correlations among symptoms. Using these multi-way probabilities in conjunction with an iterative proportional fitting algorithm allows estimation of full conditional probabilities. Combining these conditional probabilities with misdiagnosis error rates and incidence rates via Bayes theorem, the probability of each syndrome is estimated. We tested a prototype of computer-aided differential diagnosis (CADDY) on simulated data and on more than 100 real cases, including West Nile Virus, Q fever, SARS, anthrax, plague, tularaemia and toxic shock cases. We conclude that: (1) it is important to determine whether the unrecorded positive status of a symptom means that the status is negative or that the status is unknown; (2) inclusion of misdiagnosis error rates produces more realistic results; (3) the naive Bayes classifier, which assumes all symptoms behave independently, is slightly outperformed by CADDY, which includes available multi-symptom information on correlations; as more information regarding symptom correlations becomes available, the advantage of CADDY over the naive Bayes classifier should increase; (4) overlooking low-probability, high-consequence events is less likely if the standard output summary is augmented with a list of rare syndromes that are consistent with observed symptoms, and (5) accumulating patient-level probabilities across a larger population can aid in biosurveillance for disease outbreaks.</AbstractText>
<CopyrightInformation>c 2007 John Wiley & Sons, Ltd.</CopyrightInformation>
</Abstract>
<AuthorList CompleteYN="Y"><Author ValidYN="Y"><LastName>Burr</LastName>
<ForeName>Tom</ForeName>
<Initials>T</Initials>
<AffiliationInfo><Affiliation>Los Alamos National Laboratory, Mail Stop F600, Los Alamos, NM 87545, USA. tburr@lanl.gov</Affiliation>
</AffiliationInfo>
</Author>
<Author ValidYN="Y"><LastName>Koster</LastName>
<ForeName>Frederick</ForeName>
<Initials>F</Initials>
</Author>
<Author ValidYN="Y"><LastName>Picard</LastName>
<ForeName>Rick</ForeName>
<Initials>R</Initials>
</Author>
<Author ValidYN="Y"><LastName>Forslund</LastName>
<ForeName>Dave</ForeName>
<Initials>D</Initials>
</Author>
<Author ValidYN="Y"><LastName>Wokoun</LastName>
<ForeName>Doug</ForeName>
<Initials>D</Initials>
</Author>
<Author ValidYN="Y"><LastName>Joyce</LastName>
<ForeName>Ed</ForeName>
<Initials>E</Initials>
</Author>
<Author ValidYN="Y"><LastName>Brillman</LastName>
<ForeName>Judith</ForeName>
<Initials>J</Initials>
</Author>
<Author ValidYN="Y"><LastName>Froman</LastName>
<ForeName>Phil</ForeName>
<Initials>P</Initials>
</Author>
<Author ValidYN="Y"><LastName>Lee</LastName>
<ForeName>Jack</ForeName>
<Initials>J</Initials>
</Author>
</AuthorList>
<Language>eng</Language>
<PublicationTypeList><PublicationType UI="D016428">Journal Article</PublicationType>
<PublicationType UI="D013485">Research Support, Non-U.S. Gov't</PublicationType>
</PublicationTypeList>
</Article>
<MedlineJournalInfo><Country>England</Country>
<MedlineTA>Stat Med</MedlineTA>
<NlmUniqueID>8215016</NlmUniqueID>
<ISSNLinking>0277-6715</ISSNLinking>
</MedlineJournalInfo>
<CitationSubset>IM</CitationSubset>
<MeshHeadingList><MeshHeading><DescriptorName UI="D000465" MajorTopicYN="Y">Algorithms</DescriptorName>
</MeshHeading>
<MeshHeading><DescriptorName UI="D000881" MajorTopicYN="N">Anthrax</DescriptorName>
<QualifierName UI="Q000175" MajorTopicYN="N">diagnosis</QualifierName>
</MeshHeading>
<MeshHeading><DescriptorName UI="D023482" MajorTopicYN="Y">Bioterrorism</DescriptorName>
</MeshHeading>
<MeshHeading><DescriptorName UI="D003198" MajorTopicYN="N">Computer Simulation</DescriptorName>
</MeshHeading>
<MeshHeading><DescriptorName UI="D003936" MajorTopicYN="N">Diagnosis, Computer-Assisted</DescriptorName>
<QualifierName UI="Q000379" MajorTopicYN="Y">methods</QualifierName>
</MeshHeading>
<MeshHeading><DescriptorName UI="D004196" MajorTopicYN="Y">Disease Outbreaks</DescriptorName>
</MeshHeading>
<MeshHeading><DescriptorName UI="D018778" MajorTopicYN="N">Hantavirus Infections</DescriptorName>
<QualifierName UI="Q000175" MajorTopicYN="N">diagnosis</QualifierName>
</MeshHeading>
<MeshHeading><DescriptorName UI="D006801" MajorTopicYN="N">Humans</DescriptorName>
</MeshHeading>
<MeshHeading><DescriptorName UI="D012680" MajorTopicYN="N">Sensitivity and Specificity</DescriptorName>
</MeshHeading>
</MeshHeadingList>
</MedlineCitation>
<PubmedData><History><PubMedPubDate PubStatus="pubmed"><Year>2007</Year>
<Month>1</Month>
<Day>17</Day>
<Hour>9</Hour>
<Minute>0</Minute>
</PubMedPubDate>
<PubMedPubDate PubStatus="medline"><Year>2007</Year>
<Month>5</Month>
<Day>30</Day>
<Hour>9</Hour>
<Minute>0</Minute>
</PubMedPubDate>
<PubMedPubDate PubStatus="entrez"><Year>2007</Year>
<Month>1</Month>
<Day>17</Day>
<Hour>9</Hour>
<Minute>0</Minute>
</PubMedPubDate>
</History>
<PublicationStatus>ppublish</PublicationStatus>
<ArticleIdList><ArticleId IdType="pubmed">17225213</ArticleId>
<ArticleId IdType="doi">10.1002/sim.2798</ArticleId>
</ArticleIdList>
</PubmedData>
</pubmed>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Sante/explor/SrasV1/Data/PubMed/Corpus
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 001F24 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/PubMed/Corpus/biblio.hfd -nk 001F24 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Sante |area= SrasV1 |flux= PubMed |étape= Corpus |type= RBID |clé= pubmed:17225213 |texte= Computer-aided diagnosis with potential application to rapid detection of disease outbreaks. }}
Pour générer des pages wiki
HfdIndexSelect -h $EXPLOR_AREA/Data/PubMed/Corpus/RBID.i -Sk "pubmed:17225213" \ | HfdSelect -Kh $EXPLOR_AREA/Data/PubMed/Corpus/biblio.hfd \ | NlmPubMed2Wicri -a SrasV1
![]() | This area was generated with Dilib version V0.6.33. | ![]() |