Optimizing the response to surveillance alerts in automated surveillance systems
Identifieur interne : 002022 ( Istex/Curation ); précédent : 002021; suivant : 002023Optimizing the response to surveillance alerts in automated surveillance systems
Auteurs : Masoumeh Izadi [Canada] ; David L. Buckeridge [Canada]Source :
- Statistics in Medicine [ 0277-6715 ] ; 2011-02-28.
English descriptors
- Teeft :
- Action plans, Actual error, Adaptive, Adaptive optimization, Adaptive optimization approach, Additive, Additive approximation, Additive case, Additive error, Algorithm, Anthrax, Anthrax surveillance, Approximate model, Approximate pomdp model, Belief state, Biosurveillance, Buckeridge, Computer science, Conditional plan, Conditional plans, Copyright, Decision tree, Detection, Detection algorithms, Detection method, Detection methods, Detection sensitivity, Different detection methods, Discount factor, Disease outbreak, Disease outbreaks, Early detection, Empirical evaluation, Empirical results, Epidemiological response, Error bounds, Exact model, Exact pomdp model, False alarm, Future investigation, Georgia health districts, Health informatics research group, Immediate action, Immediate reward, Infectious diseases, Informatics, Intervention strategies, Investigation actions, Izadi, John wiley sons, Large number, Last step, Lecture notes, Linear factor, Markov, Markov models, Mcgill university, Medical decision, Mmwr morbidity mortality, Multiplicative, Multiplicative approximation, Multiplicative error, Multiplicative noise, Normal state, Observable markov decision processes, Observation function, Observation functions, Optimal policy, Optimal value function, Optimization, Outbreak, Outbreak detection, Outbreak detection methods, Outbreak state, Outbreak states, Parameter perturbation, Parameter values, Perturbation, Perturbation analysis, Pomdp, Pomdp approach, Pomdp model, Pomdp parameters, Pomdp policy, Pomdps, Potential costs, Preventable loss, Proof proceeds, Public health, Public health actions, Real outbreak, Results show, Review records, Reward function, Same sets, Same structure, Security informatics, Sequential decisions, Special case, State space, Statist, Statistical aberrancy detection, Such perturbations, Surveillance, Surveillance context, Surveillance data, Surveillance systems, Syndromic surveillance, Systematic studies, Systematic study, Timeliness, Total cost, Total costs, Transition function, Transition functions, True associations, True outbreaks, Utility function, Value function, Value functions, Worst case.
Abstract
Although much research effort has been directed toward refining algorithms for disease outbreak alerting, considerably less attention has been given to the response to alerts generated from statistical detection algorithms. Given the inherent inaccuracy in alerting, it is imperative to develop methods that help public health personnel identify optimal policies in response to alerts. This study evaluates the application of dynamic decision making models to the problem of responding to outbreak detection methods, using anthrax surveillance as an example. Adaptive optimization through approximate dynamic programming is used to generate a policy for decision making following outbreak detection. We investigate the degree to which the model can tolerate noise theoretically, in order to keep near optimal behavior. We also evaluate the policy from our model empirically and compare it with current approaches in routine public health practice for investigating alerts. Timeliness of outbreak confirmation and total costs associated with the decisions made are used as performance measures. Using our approach, on average, 80 per cent of outbreaks were confirmed prior to the fifth day of post‐attack with considerably less cost compared to response strategies currently in use. Experimental results are also provided to illustrate the robustness of the adaptive optimization approach and to show the realization of the derived error bounds in practice. Copyright © 2011 John Wiley & Sons, Ltd.
Url:
DOI: 10.1002/sim.3922
Links toward previous steps (curation, corpus...)
- to stream Istex, to step Corpus: Pour aller vers cette notice dans l'étape Curation :002022
Links to Exploration step
ISTEX:CF6918448731E8F7F0AA5746642D3EEADAC7FAFDLe document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title xml:lang="en">Optimizing the response to surveillance alerts in automated surveillance systems</title>
<author><name sortKey="Izadi, Masoumeh" sort="Izadi, Masoumeh" uniqKey="Izadi M" first="Masoumeh" last="Izadi">Masoumeh Izadi</name>
<affiliation wicri:level="1"><mods:affiliation>Clinical and Health Informatics Research Group, McGill University, 1140 Pine Ave., Montreal, QC, Canada H3A 1A3</mods:affiliation>
<country>Canada</country>
<wicri:regionArea>Clinical and Health Informatics Research Group, McGill University, 1140 Pine Ave., Montreal, QC</wicri:regionArea>
</affiliation>
<affiliation wicri:level="1"><mods:affiliation>E-mail: mtabae@cs.mcgill.ca</mods:affiliation>
<country wicri:rule="url">Canada</country>
</affiliation>
<affiliation wicri:level="1"><mods:affiliation>Correspondence address: Clinical and Health Informatics Research Group, McGill University, 1140 Pine Ave., Montreal, QC, Canada H3A 1A3.</mods:affiliation>
<country>Canada</country>
<wicri:regionArea>Correspondence address: Clinical and Health Informatics Research Group, McGill University, 1140 Pine Ave., Montreal, QC</wicri:regionArea>
</affiliation>
</author>
<author><name sortKey="Buckeridge, David L" sort="Buckeridge, David L" uniqKey="Buckeridge D" first="David L." last="Buckeridge">David L. Buckeridge</name>
<affiliation wicri:level="1"><mods:affiliation>Clinical and Health Informatics Research Group, McGill University, 1140 Pine Ave., Montreal, QC, Canada H3A 1A3</mods:affiliation>
<country>Canada</country>
<wicri:regionArea>Clinical and Health Informatics Research Group, McGill University, 1140 Pine Ave., Montreal, QC</wicri:regionArea>
</affiliation>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">ISTEX</idno>
<idno type="RBID">ISTEX:CF6918448731E8F7F0AA5746642D3EEADAC7FAFD</idno>
<date when="2011" year="2011">2011</date>
<idno type="doi">10.1002/sim.3922</idno>
<idno type="url">https://api.istex.fr/ark:/67375/WNG-CFNHTM13-9/fulltext.pdf</idno>
<idno type="wicri:Area/Istex/Corpus">002022</idno>
<idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Corpus" wicri:corpus="ISTEX">002022</idno>
<idno type="wicri:Area/Istex/Curation">002022</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title level="a" type="main">Optimizing the response to surveillance alerts in automated surveillance systems</title>
<author><name sortKey="Izadi, Masoumeh" sort="Izadi, Masoumeh" uniqKey="Izadi M" first="Masoumeh" last="Izadi">Masoumeh Izadi</name>
<affiliation wicri:level="1"><mods:affiliation>Clinical and Health Informatics Research Group, McGill University, 1140 Pine Ave., Montreal, QC, Canada H3A 1A3</mods:affiliation>
<country>Canada</country>
<wicri:regionArea>Clinical and Health Informatics Research Group, McGill University, 1140 Pine Ave., Montreal, QC</wicri:regionArea>
</affiliation>
<affiliation wicri:level="1"><mods:affiliation>E-mail: mtabae@cs.mcgill.ca</mods:affiliation>
<country wicri:rule="url">Canada</country>
</affiliation>
<affiliation wicri:level="1"><mods:affiliation>Correspondence address: Clinical and Health Informatics Research Group, McGill University, 1140 Pine Ave., Montreal, QC, Canada H3A 1A3.</mods:affiliation>
<country>Canada</country>
<wicri:regionArea>Correspondence address: Clinical and Health Informatics Research Group, McGill University, 1140 Pine Ave., Montreal, QC</wicri:regionArea>
</affiliation>
</author>
<author><name sortKey="Buckeridge, David L" sort="Buckeridge, David L" uniqKey="Buckeridge D" first="David L." last="Buckeridge">David L. Buckeridge</name>
<affiliation wicri:level="1"><mods:affiliation>Clinical and Health Informatics Research Group, McGill University, 1140 Pine Ave., Montreal, QC, Canada H3A 1A3</mods:affiliation>
<country>Canada</country>
<wicri:regionArea>Clinical and Health Informatics Research Group, McGill University, 1140 Pine Ave., Montreal, QC</wicri:regionArea>
</affiliation>
</author>
</analytic>
<monogr></monogr>
<series><title level="j" type="main">Statistics in Medicine</title>
<title level="j" type="sub">13th Biennial CDC & ATSDR Symposium on Statistical Methods Info‐Fusion: Utilization of Multi‐Source Data</title>
<title level="j" type="alt">STATISTICS IN MEDICINE</title>
<idno type="ISSN">0277-6715</idno>
<idno type="eISSN">1097-0258</idno>
<imprint><biblScope unit="vol">30</biblScope>
<biblScope unit="issue">5</biblScope>
<biblScope unit="page" from="442">442</biblScope>
<biblScope unit="page" to="454">454</biblScope>
<biblScope unit="page-count">13</biblScope>
<publisher>John Wiley & Sons, Ltd.</publisher>
<pubPlace>Chichester, UK</pubPlace>
<date type="published" when="2011-02-28">2011-02-28</date>
</imprint>
<idno type="ISSN">0277-6715</idno>
</series>
</biblStruct>
</sourceDesc>
<seriesStmt><idno type="ISSN">0277-6715</idno>
</seriesStmt>
</fileDesc>
<profileDesc><textClass><keywords scheme="Teeft" xml:lang="en"><term>Action plans</term>
<term>Actual error</term>
<term>Adaptive</term>
<term>Adaptive optimization</term>
<term>Adaptive optimization approach</term>
<term>Additive</term>
<term>Additive approximation</term>
<term>Additive case</term>
<term>Additive error</term>
<term>Algorithm</term>
<term>Anthrax</term>
<term>Anthrax surveillance</term>
<term>Approximate model</term>
<term>Approximate pomdp model</term>
<term>Belief state</term>
<term>Biosurveillance</term>
<term>Buckeridge</term>
<term>Computer science</term>
<term>Conditional plan</term>
<term>Conditional plans</term>
<term>Copyright</term>
<term>Decision tree</term>
<term>Detection</term>
<term>Detection algorithms</term>
<term>Detection method</term>
<term>Detection methods</term>
<term>Detection sensitivity</term>
<term>Different detection methods</term>
<term>Discount factor</term>
<term>Disease outbreak</term>
<term>Disease outbreaks</term>
<term>Early detection</term>
<term>Empirical evaluation</term>
<term>Empirical results</term>
<term>Epidemiological response</term>
<term>Error bounds</term>
<term>Exact model</term>
<term>Exact pomdp model</term>
<term>False alarm</term>
<term>Future investigation</term>
<term>Georgia health districts</term>
<term>Health informatics research group</term>
<term>Immediate action</term>
<term>Immediate reward</term>
<term>Infectious diseases</term>
<term>Informatics</term>
<term>Intervention strategies</term>
<term>Investigation actions</term>
<term>Izadi</term>
<term>John wiley sons</term>
<term>Large number</term>
<term>Last step</term>
<term>Lecture notes</term>
<term>Linear factor</term>
<term>Markov</term>
<term>Markov models</term>
<term>Mcgill university</term>
<term>Medical decision</term>
<term>Mmwr morbidity mortality</term>
<term>Multiplicative</term>
<term>Multiplicative approximation</term>
<term>Multiplicative error</term>
<term>Multiplicative noise</term>
<term>Normal state</term>
<term>Observable markov decision processes</term>
<term>Observation function</term>
<term>Observation functions</term>
<term>Optimal policy</term>
<term>Optimal value function</term>
<term>Optimization</term>
<term>Outbreak</term>
<term>Outbreak detection</term>
<term>Outbreak detection methods</term>
<term>Outbreak state</term>
<term>Outbreak states</term>
<term>Parameter perturbation</term>
<term>Parameter values</term>
<term>Perturbation</term>
<term>Perturbation analysis</term>
<term>Pomdp</term>
<term>Pomdp approach</term>
<term>Pomdp model</term>
<term>Pomdp parameters</term>
<term>Pomdp policy</term>
<term>Pomdps</term>
<term>Potential costs</term>
<term>Preventable loss</term>
<term>Proof proceeds</term>
<term>Public health</term>
<term>Public health actions</term>
<term>Real outbreak</term>
<term>Results show</term>
<term>Review records</term>
<term>Reward function</term>
<term>Same sets</term>
<term>Same structure</term>
<term>Security informatics</term>
<term>Sequential decisions</term>
<term>Special case</term>
<term>State space</term>
<term>Statist</term>
<term>Statistical aberrancy detection</term>
<term>Such perturbations</term>
<term>Surveillance</term>
<term>Surveillance context</term>
<term>Surveillance data</term>
<term>Surveillance systems</term>
<term>Syndromic surveillance</term>
<term>Systematic studies</term>
<term>Systematic study</term>
<term>Timeliness</term>
<term>Total cost</term>
<term>Total costs</term>
<term>Transition function</term>
<term>Transition functions</term>
<term>True associations</term>
<term>True outbreaks</term>
<term>Utility function</term>
<term>Value function</term>
<term>Value functions</term>
<term>Worst case</term>
</keywords>
</textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">Although much research effort has been directed toward refining algorithms for disease outbreak alerting, considerably less attention has been given to the response to alerts generated from statistical detection algorithms. Given the inherent inaccuracy in alerting, it is imperative to develop methods that help public health personnel identify optimal policies in response to alerts. This study evaluates the application of dynamic decision making models to the problem of responding to outbreak detection methods, using anthrax surveillance as an example. Adaptive optimization through approximate dynamic programming is used to generate a policy for decision making following outbreak detection. We investigate the degree to which the model can tolerate noise theoretically, in order to keep near optimal behavior. We also evaluate the policy from our model empirically and compare it with current approaches in routine public health practice for investigating alerts. Timeliness of outbreak confirmation and total costs associated with the decisions made are used as performance measures. Using our approach, on average, 80 per cent of outbreaks were confirmed prior to the fifth day of post‐attack with considerably less cost compared to response strategies currently in use. Experimental results are also provided to illustrate the robustness of the adaptive optimization approach and to show the realization of the derived error bounds in practice. Copyright © 2011 John Wiley & Sons, Ltd.</div>
</front>
</TEI>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Sante/explor/SrasV1/Data/Istex/Curation
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 002022 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Istex/Curation/biblio.hfd -nk 002022 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Sante |area= SrasV1 |flux= Istex |étape= Curation |type= RBID |clé= ISTEX:CF6918448731E8F7F0AA5746642D3EEADAC7FAFD |texte= Optimizing the response to surveillance alerts in automated surveillance systems }}
This area was generated with Dilib version V0.6.33. |