Estimation of pulmonary artery occlusion pressure by an artificial neural network
OBJECTIVEWe hypothesized that an artificial neural network, interconnected computer elements capable of adaptation and learning, could accurately estimate pulmonary artery occlusion pressure from the pulsatile pulmonary artery waveform. SETTINGUniversity medical center. SUBJECTSNineteen closed-chest...
Gespeichert in:
| Veröffentlicht in: | Critical care medicine 2003-01, Vol.31 (1), p.261-266 |
|---|---|
| Hauptverfasser: | , , , , , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 266 |
|---|---|
| container_issue | 1 |
| container_start_page | 261 |
| container_title | Critical care medicine |
| container_volume | 31 |
| creator | deBoisblanc, Bennett P Pellett, Andrew Johnson, Royce Champagne, Michael McClarty, Espisito Dhillon, Gundeep Levitzky, Michael |
| description | OBJECTIVEWe hypothesized that an artificial neural network, interconnected computer elements capable of adaptation and learning, could accurately estimate pulmonary artery occlusion pressure from the pulsatile pulmonary artery waveform.
SETTINGUniversity medical center.
SUBJECTSNineteen closed-chest dogs.
INTERVENTIONSPulmonary artery waveforms were digitally sampled before conventional measurements of pulmonary artery occlusion pressure under control conditions, during infusions of serotonin or histamine, or during volume loading. Individual beats were parsed or separated out. Pulmonary artery pressure, its first time derivative, and the beat duration were used as neural inputs. The neural network was trained by using 80% of all samples and tested on the remaining 20%. For comparison, the regression between pulmonary artery diastolic pressure and pulmonary artery occlusion pressure was developed and tested using the same data sets. As a final test of generalizability, the neural network was trained on data obtained from 18 dogs and tested on data from the remaining dog in a round-robin fashion.
MEASUREMENTS AND MAIN RESULTSThe correlation coefficient between the pulmonary artery diastolic pressure estimate of pulmonary artery occlusion pressure and measured pulmonary artery occlusion pressure was .75, whereas that for the neural network estimate of pulmonary artery occlusion pressure was .97 (p < .01 for difference between pulmonary artery diastolic pressure and pulmonary artery occlusion pressure estimates). The pulmonary artery diastolic pressure estimate of pulmonary artery occlusion pressure showed a bias of 0.097 mm Hg (limits of agreement −7.57 to 7.767 mm Hg), whereas the neural network estimate of pulmonary artery occlusion pressure showed a bias of −0.002 mm Hg (−2.592 to 2.588 mm Hg). There was no significant change in the bias of the neural network estimate over the range of values tested. In contrast, the bias for the pulmonary artery diastolic pressure estimate significantly increased with the increasing magnitude of the pulmonary artery occlusion pressure. During round-robin testing, the neural network estimate of pulmonary artery occlusion pressure showed suboptimal performance (correlation coefficient between estimated and measured pulmonary artery occlusion pressure .59).
CONCLUSIONSA neural network can accurately estimate pulmonary artery occlusion pressure over a wide range of pulmonary artery occlusion pressure under conditions that al |
| doi_str_mv | 10.1097/00003246-200301000-00041 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_miscellaneous_72996950</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>72996950</sourcerecordid><originalsourceid>FETCH-LOGICAL-c4521-1b0b7f27e6185029faca0a8e8351be43fe18dc5863f49739dcfaeac0346fc12b3</originalsourceid><addsrcrecordid>eNp1kV1LwzAUhoMobk7_gvRG76r5aptcypgfMBBBr0OanbC6rJlJy9i_N9uqXhkIL4f3OSfJG4Qygu8IltU9TotRXuY0KSapytPm5ASNScFSQSU7RWOMJc4Zl2yELmL8xJjwomLnaERowQtMyzF6m8WuWeuu8W3mbbbp3dq3OuwyHTpI4o1xfdy7mwAx9gGyOpnt3m9sYxrtshb6cJBu68PqEp1Z7SJcDTpBH4-z9-lzPn99epk-zHPDC0pyUuO6srSCkoh0FWm10VgLEKwgNXBmgYiFKUTJLJcVkwtjNWiDGS-tIbRmE3R7nLsJ_quH2Kl1Ew04p1vwfVQVlbKUBU6gOIIm-BgDWLUJ6clhpwhW-zjVT5zqN051iDO1Xg9n9PUaFn-NQ34JuBkAHY12NujWNPGP41wwIXji-JHbepdyjSvXbyGoJWjXLdV_38m-ARcKjXQ</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>72996950</pqid></control><display><type>article</type><title>Estimation of pulmonary artery occlusion pressure by an artificial neural network</title><source>MEDLINE</source><source>Journals@Ovid Ovid Autoload</source><creator>deBoisblanc, Bennett P ; Pellett, Andrew ; Johnson, Royce ; Champagne, Michael ; McClarty, Espisito ; Dhillon, Gundeep ; Levitzky, Michael</creator><creatorcontrib>deBoisblanc, Bennett P ; Pellett, Andrew ; Johnson, Royce ; Champagne, Michael ; McClarty, Espisito ; Dhillon, Gundeep ; Levitzky, Michael</creatorcontrib><description>OBJECTIVEWe hypothesized that an artificial neural network, interconnected computer elements capable of adaptation and learning, could accurately estimate pulmonary artery occlusion pressure from the pulsatile pulmonary artery waveform.
SETTINGUniversity medical center.
SUBJECTSNineteen closed-chest dogs.
INTERVENTIONSPulmonary artery waveforms were digitally sampled before conventional measurements of pulmonary artery occlusion pressure under control conditions, during infusions of serotonin or histamine, or during volume loading. Individual beats were parsed or separated out. Pulmonary artery pressure, its first time derivative, and the beat duration were used as neural inputs. The neural network was trained by using 80% of all samples and tested on the remaining 20%. For comparison, the regression between pulmonary artery diastolic pressure and pulmonary artery occlusion pressure was developed and tested using the same data sets. As a final test of generalizability, the neural network was trained on data obtained from 18 dogs and tested on data from the remaining dog in a round-robin fashion.
MEASUREMENTS AND MAIN RESULTSThe correlation coefficient between the pulmonary artery diastolic pressure estimate of pulmonary artery occlusion pressure and measured pulmonary artery occlusion pressure was .75, whereas that for the neural network estimate of pulmonary artery occlusion pressure was .97 (p < .01 for difference between pulmonary artery diastolic pressure and pulmonary artery occlusion pressure estimates). The pulmonary artery diastolic pressure estimate of pulmonary artery occlusion pressure showed a bias of 0.097 mm Hg (limits of agreement −7.57 to 7.767 mm Hg), whereas the neural network estimate of pulmonary artery occlusion pressure showed a bias of −0.002 mm Hg (−2.592 to 2.588 mm Hg). There was no significant change in the bias of the neural network estimate over the range of values tested. In contrast, the bias for the pulmonary artery diastolic pressure estimate significantly increased with the increasing magnitude of the pulmonary artery occlusion pressure. During round-robin testing, the neural network estimate of pulmonary artery occlusion pressure showed suboptimal performance (correlation coefficient between estimated and measured pulmonary artery occlusion pressure .59).
CONCLUSIONSA neural network can accurately estimate pulmonary artery occlusion pressure over a wide range of pulmonary artery occlusion pressure under conditions that alter pulmonary hemodynamics. We speculate that artificial neural networks could provide accurate, real-time estimates of pulmonary artery occlusion pressure in critically ill patients.</description><identifier>ISSN: 0090-3493</identifier><identifier>EISSN: 1530-0293</identifier><identifier>DOI: 10.1097/00003246-200301000-00041</identifier><identifier>PMID: 12545026</identifier><identifier>CODEN: CCMDC7</identifier><language>eng</language><publisher>Hagerstown, MD: by the Society of Critical Care Medicine and Lippincott Williams & Wilkins</publisher><subject>Animals ; Biological and medical sciences ; Blood Pressure Determination - methods ; Cardiovascular system ; Diagnosis, Computer-Assisted - methods ; Dogs ; Hemodynamics ; Investigative techniques of hemodynamics ; Investigative techniques, diagnostic techniques (general aspects) ; Linear Models ; Medical sciences ; Neural Networks (Computer) ; Pulmonary Wedge Pressure ; Signal Processing, Computer-Assisted</subject><ispartof>Critical care medicine, 2003-01, Vol.31 (1), p.261-266</ispartof><rights>2003 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins</rights><rights>2003 INIST-CNRS</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c4521-1b0b7f27e6185029faca0a8e8351be43fe18dc5863f49739dcfaeac0346fc12b3</citedby><cites>FETCH-LOGICAL-c4521-1b0b7f27e6185029faca0a8e8351be43fe18dc5863f49739dcfaeac0346fc12b3</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,780,784,4024,27923,27924,27925</link.rule.ids><backlink>$$Uhttp://pascal-francis.inist.fr/vibad/index.php?action=getRecordDetail&idt=14483884$$DView record in Pascal Francis$$Hfree_for_read</backlink><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/12545026$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>deBoisblanc, Bennett P</creatorcontrib><creatorcontrib>Pellett, Andrew</creatorcontrib><creatorcontrib>Johnson, Royce</creatorcontrib><creatorcontrib>Champagne, Michael</creatorcontrib><creatorcontrib>McClarty, Espisito</creatorcontrib><creatorcontrib>Dhillon, Gundeep</creatorcontrib><creatorcontrib>Levitzky, Michael</creatorcontrib><title>Estimation of pulmonary artery occlusion pressure by an artificial neural network</title><title>Critical care medicine</title><addtitle>Crit Care Med</addtitle><description>OBJECTIVEWe hypothesized that an artificial neural network, interconnected computer elements capable of adaptation and learning, could accurately estimate pulmonary artery occlusion pressure from the pulsatile pulmonary artery waveform.
SETTINGUniversity medical center.
SUBJECTSNineteen closed-chest dogs.
INTERVENTIONSPulmonary artery waveforms were digitally sampled before conventional measurements of pulmonary artery occlusion pressure under control conditions, during infusions of serotonin or histamine, or during volume loading. Individual beats were parsed or separated out. Pulmonary artery pressure, its first time derivative, and the beat duration were used as neural inputs. The neural network was trained by using 80% of all samples and tested on the remaining 20%. For comparison, the regression between pulmonary artery diastolic pressure and pulmonary artery occlusion pressure was developed and tested using the same data sets. As a final test of generalizability, the neural network was trained on data obtained from 18 dogs and tested on data from the remaining dog in a round-robin fashion.
MEASUREMENTS AND MAIN RESULTSThe correlation coefficient between the pulmonary artery diastolic pressure estimate of pulmonary artery occlusion pressure and measured pulmonary artery occlusion pressure was .75, whereas that for the neural network estimate of pulmonary artery occlusion pressure was .97 (p < .01 for difference between pulmonary artery diastolic pressure and pulmonary artery occlusion pressure estimates). The pulmonary artery diastolic pressure estimate of pulmonary artery occlusion pressure showed a bias of 0.097 mm Hg (limits of agreement −7.57 to 7.767 mm Hg), whereas the neural network estimate of pulmonary artery occlusion pressure showed a bias of −0.002 mm Hg (−2.592 to 2.588 mm Hg). There was no significant change in the bias of the neural network estimate over the range of values tested. In contrast, the bias for the pulmonary artery diastolic pressure estimate significantly increased with the increasing magnitude of the pulmonary artery occlusion pressure. During round-robin testing, the neural network estimate of pulmonary artery occlusion pressure showed suboptimal performance (correlation coefficient between estimated and measured pulmonary artery occlusion pressure .59).
CONCLUSIONSA neural network can accurately estimate pulmonary artery occlusion pressure over a wide range of pulmonary artery occlusion pressure under conditions that alter pulmonary hemodynamics. We speculate that artificial neural networks could provide accurate, real-time estimates of pulmonary artery occlusion pressure in critically ill patients.</description><subject>Animals</subject><subject>Biological and medical sciences</subject><subject>Blood Pressure Determination - methods</subject><subject>Cardiovascular system</subject><subject>Diagnosis, Computer-Assisted - methods</subject><subject>Dogs</subject><subject>Hemodynamics</subject><subject>Investigative techniques of hemodynamics</subject><subject>Investigative techniques, diagnostic techniques (general aspects)</subject><subject>Linear Models</subject><subject>Medical sciences</subject><subject>Neural Networks (Computer)</subject><subject>Pulmonary Wedge Pressure</subject><subject>Signal Processing, Computer-Assisted</subject><issn>0090-3493</issn><issn>1530-0293</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2003</creationdate><recordtype>article</recordtype><sourceid>EIF</sourceid><recordid>eNp1kV1LwzAUhoMobk7_gvRG76r5aptcypgfMBBBr0OanbC6rJlJy9i_N9uqXhkIL4f3OSfJG4Qygu8IltU9TotRXuY0KSapytPm5ASNScFSQSU7RWOMJc4Zl2yELmL8xJjwomLnaERowQtMyzF6m8WuWeuu8W3mbbbp3dq3OuwyHTpI4o1xfdy7mwAx9gGyOpnt3m9sYxrtshb6cJBu68PqEp1Z7SJcDTpBH4-z9-lzPn99epk-zHPDC0pyUuO6srSCkoh0FWm10VgLEKwgNXBmgYiFKUTJLJcVkwtjNWiDGS-tIbRmE3R7nLsJ_quH2Kl1Ew04p1vwfVQVlbKUBU6gOIIm-BgDWLUJ6clhpwhW-zjVT5zqN051iDO1Xg9n9PUaFn-NQ34JuBkAHY12NujWNPGP41wwIXji-JHbepdyjSvXbyGoJWjXLdV_38m-ARcKjXQ</recordid><startdate>200301</startdate><enddate>200301</enddate><creator>deBoisblanc, Bennett P</creator><creator>Pellett, Andrew</creator><creator>Johnson, Royce</creator><creator>Champagne, Michael</creator><creator>McClarty, Espisito</creator><creator>Dhillon, Gundeep</creator><creator>Levitzky, Michael</creator><general>by the Society of Critical Care Medicine and Lippincott Williams & Wilkins</general><general>Lippincott</general><scope>IQODW</scope><scope>CGR</scope><scope>CUY</scope><scope>CVF</scope><scope>ECM</scope><scope>EIF</scope><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope></search><sort><creationdate>200301</creationdate><title>Estimation of pulmonary artery occlusion pressure by an artificial neural network</title><author>deBoisblanc, Bennett P ; Pellett, Andrew ; Johnson, Royce ; Champagne, Michael ; McClarty, Espisito ; Dhillon, Gundeep ; Levitzky, Michael</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c4521-1b0b7f27e6185029faca0a8e8351be43fe18dc5863f49739dcfaeac0346fc12b3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2003</creationdate><topic>Animals</topic><topic>Biological and medical sciences</topic><topic>Blood Pressure Determination - methods</topic><topic>Cardiovascular system</topic><topic>Diagnosis, Computer-Assisted - methods</topic><topic>Dogs</topic><topic>Hemodynamics</topic><topic>Investigative techniques of hemodynamics</topic><topic>Investigative techniques, diagnostic techniques (general aspects)</topic><topic>Linear Models</topic><topic>Medical sciences</topic><topic>Neural Networks (Computer)</topic><topic>Pulmonary Wedge Pressure</topic><topic>Signal Processing, Computer-Assisted</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>deBoisblanc, Bennett P</creatorcontrib><creatorcontrib>Pellett, Andrew</creatorcontrib><creatorcontrib>Johnson, Royce</creatorcontrib><creatorcontrib>Champagne, Michael</creatorcontrib><creatorcontrib>McClarty, Espisito</creatorcontrib><creatorcontrib>Dhillon, Gundeep</creatorcontrib><creatorcontrib>Levitzky, Michael</creatorcontrib><collection>Pascal-Francis</collection><collection>Medline</collection><collection>MEDLINE</collection><collection>MEDLINE (Ovid)</collection><collection>MEDLINE</collection><collection>MEDLINE</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><jtitle>Critical care medicine</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>deBoisblanc, Bennett P</au><au>Pellett, Andrew</au><au>Johnson, Royce</au><au>Champagne, Michael</au><au>McClarty, Espisito</au><au>Dhillon, Gundeep</au><au>Levitzky, Michael</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Estimation of pulmonary artery occlusion pressure by an artificial neural network</atitle><jtitle>Critical care medicine</jtitle><addtitle>Crit Care Med</addtitle><date>2003-01</date><risdate>2003</risdate><volume>31</volume><issue>1</issue><spage>261</spage><epage>266</epage><pages>261-266</pages><issn>0090-3493</issn><eissn>1530-0293</eissn><coden>CCMDC7</coden><abstract>OBJECTIVEWe hypothesized that an artificial neural network, interconnected computer elements capable of adaptation and learning, could accurately estimate pulmonary artery occlusion pressure from the pulsatile pulmonary artery waveform.
SETTINGUniversity medical center.
SUBJECTSNineteen closed-chest dogs.
INTERVENTIONSPulmonary artery waveforms were digitally sampled before conventional measurements of pulmonary artery occlusion pressure under control conditions, during infusions of serotonin or histamine, or during volume loading. Individual beats were parsed or separated out. Pulmonary artery pressure, its first time derivative, and the beat duration were used as neural inputs. The neural network was trained by using 80% of all samples and tested on the remaining 20%. For comparison, the regression between pulmonary artery diastolic pressure and pulmonary artery occlusion pressure was developed and tested using the same data sets. As a final test of generalizability, the neural network was trained on data obtained from 18 dogs and tested on data from the remaining dog in a round-robin fashion.
MEASUREMENTS AND MAIN RESULTSThe correlation coefficient between the pulmonary artery diastolic pressure estimate of pulmonary artery occlusion pressure and measured pulmonary artery occlusion pressure was .75, whereas that for the neural network estimate of pulmonary artery occlusion pressure was .97 (p < .01 for difference between pulmonary artery diastolic pressure and pulmonary artery occlusion pressure estimates). The pulmonary artery diastolic pressure estimate of pulmonary artery occlusion pressure showed a bias of 0.097 mm Hg (limits of agreement −7.57 to 7.767 mm Hg), whereas the neural network estimate of pulmonary artery occlusion pressure showed a bias of −0.002 mm Hg (−2.592 to 2.588 mm Hg). There was no significant change in the bias of the neural network estimate over the range of values tested. In contrast, the bias for the pulmonary artery diastolic pressure estimate significantly increased with the increasing magnitude of the pulmonary artery occlusion pressure. During round-robin testing, the neural network estimate of pulmonary artery occlusion pressure showed suboptimal performance (correlation coefficient between estimated and measured pulmonary artery occlusion pressure .59).
CONCLUSIONSA neural network can accurately estimate pulmonary artery occlusion pressure over a wide range of pulmonary artery occlusion pressure under conditions that alter pulmonary hemodynamics. We speculate that artificial neural networks could provide accurate, real-time estimates of pulmonary artery occlusion pressure in critically ill patients.</abstract><cop>Hagerstown, MD</cop><pub>by the Society of Critical Care Medicine and Lippincott Williams & Wilkins</pub><pmid>12545026</pmid><doi>10.1097/00003246-200301000-00041</doi><tpages>6</tpages></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 0090-3493 |
| ispartof | Critical care medicine, 2003-01, Vol.31 (1), p.261-266 |
| issn | 0090-3493 1530-0293 |
| language | eng |
| recordid | cdi_proquest_miscellaneous_72996950 |
| source | MEDLINE; Journals@Ovid Ovid Autoload |
| subjects | Animals Biological and medical sciences Blood Pressure Determination - methods Cardiovascular system Diagnosis, Computer-Assisted - methods Dogs Hemodynamics Investigative techniques of hemodynamics Investigative techniques, diagnostic techniques (general aspects) Linear Models Medical sciences Neural Networks (Computer) Pulmonary Wedge Pressure Signal Processing, Computer-Assisted |
| title | Estimation of pulmonary artery occlusion pressure by an artificial neural network |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-08T05%3A27%3A11IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Estimation%20of%20pulmonary%20artery%20occlusion%20pressure%20by%20an%20artificial%20neural%20network&rft.jtitle=Critical%20care%20medicine&rft.au=deBoisblanc,%20Bennett%20P&rft.date=2003-01&rft.volume=31&rft.issue=1&rft.spage=261&rft.epage=266&rft.pages=261-266&rft.issn=0090-3493&rft.eissn=1530-0293&rft.coden=CCMDC7&rft_id=info:doi/10.1097/00003246-200301000-00041&rft_dat=%3Cproquest_cross%3E72996950%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=72996950&rft_id=info:pmid/12545026&rfr_iscdi=true |