29Prognostic safety of automatic cancellation of rest myocardial perfusion scan by machine learning: a report from multicenter REFINE SPECT registry of new generation SPECT
Abstract Background We aimed to develop a machine learning (ML) computer score derived from stress imaging and clinical data, which indicates if the rest scan could be automatically and safely canceled in the routine stress/rest myocardial perfusion SPECT (MPS). Methods A total of 20414 stress/rest...
Gespeichert in:
| Veröffentlicht in: | European heart journal 2019-10, Vol.40 (Supplement_1) |
|---|---|
| Hauptverfasser: | , , , , , , , , , , , , , |
| Format: | Artikel |
| Sprache: | eng |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | |
|---|---|
| container_issue | Supplement_1 |
| container_start_page | |
| container_title | European heart journal |
| container_volume | 40 |
| creator | Hu, L Sharir, T Fish, M B Ruddy, T D Di Carli, M Dorbala, S Einstein, A J Betancur, J Eisenberg, E Commandeur, F Germano, G Damini, D Berman, D Slomka, P J |
| description | Abstract
Background
We aimed to develop a machine learning (ML) computer score derived from stress imaging and clinical data, which indicates if the rest scan could be automatically and safely canceled in the routine stress/rest myocardial perfusion SPECT (MPS).
Methods
A total of 20414 stress/rest cases from the REFINE SPECT registry collected from 5 sites in 3 countries with Tc-99m-based MPS images, clinical data, and clinical follow-up were included in the study. All images were automatically processed at our Medical Center. The automatically generated myocardial contours were checked by experienced technologists. In total, 93 variables (26 clinical, 17 stress-test, and 50 stress-imaging variables) were used to build a LogitBoost model for prediction of adverse events (AE), including coronary revascularization, death, myocardial infarction, and unstable angina. 10-fold cross-validation was performed to separate test from validation data for the assessment of ML. The overall ML predictive performance was compared to quantitative (stress total perfusion deficit [TPD]) by the area under the receiver operating characteristic curves (AUC). ML cut-off (ML1) to simulate the decision of cancellation of the rest scan was set to result in the same % of normal scans as these determined by the normal clinical reader diagnosis on a 4-point scale in the whole population, or the same % of scans with visual summed stress scores (SSS) = 0 in the subpopulation with available SSS. A second ML cutoff (ML2) was established to achieve a 1% annual risk of AE. The annual risk of AE of the normal ML score was compared with normal clinical diagnosis and with the finding of SSS = 0.
Results
The mean follow-up interval was 4.7±1.5 years. Overall, 3542 AE were observed (3.7% annual risk). The AUC for AE was higher for ML (0.780±0.005) than for stress TPD (0.698±0.006) (p |
| doi_str_mv | 10.1093/eurheartj/ehz747.0001 |
| format | Article |
| fullrecord | <record><control><sourceid>oup_cross</sourceid><recordid>TN_cdi_crossref_primary_10_1093_eurheartj_ehz747_0001</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><oup_id>10.1093/eurheartj/ehz747.0001</oup_id><sourcerecordid>10.1093/eurheartj/ehz747.0001</sourcerecordid><originalsourceid>FETCH-LOGICAL-c1391-c1cefe2e4cfc2d4a83cf8517a2c9d2e22b8e4b91b8890761654f4bbe5efc27a13</originalsourceid><addsrcrecordid>eNqNkN9KwzAUxoMoOKePIOQFuiXpf-9kbDoYKjrBu5JmJ11Hm5SkRerT-AK-hE9muorX3pwccr7vO4cfQteUzChJ_Tl0Zg_ctIc57D_iIJ4RQugJmtCQMS-NgvAUTQhNQy-KkrdzdGHtwSmSiEYT9MXSJ6MLpW1bCmy5hLbHWmLetbrmw5_gSkBVuV6rYWLAtrjuteBmV_IKN2BkZ4ehdVKc97jmYl8qwJU7SpWquMH8-9NAo02LpdE1rrvKJYNqweDn5Wr9sMQvT8vF1mUXpW3N8QIF77gABWbcfBRcojPJKwtXv-8Uva6W28W9t3m8Wy9uN56gfkpdFSCBQSCkYLuAJ76QSUhjzkS6Y8BYnkCQpzRPkpTEjkMYyCDPIQSnjzn1pygcc4XR1hqQWWPKmps-oyQbmGd_zLOReTYwdz4y-nTX_NPyA-fwjtU</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>29Prognostic safety of automatic cancellation of rest myocardial perfusion scan by machine learning: a report from multicenter REFINE SPECT registry of new generation SPECT</title><source>Elektronische Zeitschriftenbibliothek - Frei zugängliche E-Journals</source><source>Oxford University Press Journals All Titles (1996-Current)</source><source>Alma/SFX Local Collection</source><creator>Hu, L ; Sharir, T ; Fish, M B ; Ruddy, T D ; Di Carli, M ; Dorbala, S ; Einstein, A J ; Betancur, J ; Eisenberg, E ; Commandeur, F ; Germano, G ; Damini, D ; Berman, D ; Slomka, P J</creator><creatorcontrib>Hu, L ; Sharir, T ; Fish, M B ; Ruddy, T D ; Di Carli, M ; Dorbala, S ; Einstein, A J ; Betancur, J ; Eisenberg, E ; Commandeur, F ; Germano, G ; Damini, D ; Berman, D ; Slomka, P J</creatorcontrib><description>Abstract
Background
We aimed to develop a machine learning (ML) computer score derived from stress imaging and clinical data, which indicates if the rest scan could be automatically and safely canceled in the routine stress/rest myocardial perfusion SPECT (MPS).
Methods
A total of 20414 stress/rest cases from the REFINE SPECT registry collected from 5 sites in 3 countries with Tc-99m-based MPS images, clinical data, and clinical follow-up were included in the study. All images were automatically processed at our Medical Center. The automatically generated myocardial contours were checked by experienced technologists. In total, 93 variables (26 clinical, 17 stress-test, and 50 stress-imaging variables) were used to build a LogitBoost model for prediction of adverse events (AE), including coronary revascularization, death, myocardial infarction, and unstable angina. 10-fold cross-validation was performed to separate test from validation data for the assessment of ML. The overall ML predictive performance was compared to quantitative (stress total perfusion deficit [TPD]) by the area under the receiver operating characteristic curves (AUC). ML cut-off (ML1) to simulate the decision of cancellation of the rest scan was set to result in the same % of normal scans as these determined by the normal clinical reader diagnosis on a 4-point scale in the whole population, or the same % of scans with visual summed stress scores (SSS) = 0 in the subpopulation with available SSS. A second ML cutoff (ML2) was established to achieve a 1% annual risk of AE. The annual risk of AE of the normal ML score was compared with normal clinical diagnosis and with the finding of SSS = 0.
Results
The mean follow-up interval was 4.7±1.5 years. Overall, 3542 AE were observed (3.7% annual risk). The AUC for AE was higher for ML (0.780±0.005) than for stress TPD (0.698±0.006) (p<0.001). Normal clinical diagnosis was reported in 60% cases. In 70% (14242 scans) with available segmental scores, 53% had SSS=0. ML1 and ML2 thresholds were compared with normal visual diagnosis and with SSS = 0 for AE (Figure). ML1 achieved a lower annual risk (1.5%) than normal clinical diagnosis (2.1%) or SSS = 0 (1.6% versus 2.3%) (p<0.001). The more conservative ML2 threshold with a 1% annual risk of AE resulted in a 40% canceling rate.
Figure 1
Conclusion
ML could be used to automatically cancel the rest MPS scan with the same proportion as using normal visual MPS reading, but with significantly lower AE rate in stress-only scans.
Acknowledgement/Funding
R01HL089765 from the National Heart, Lung, and Blood Institute/National Institutes of Health (NHLBI/NIH)</description><identifier>ISSN: 0195-668X</identifier><identifier>EISSN: 1522-9645</identifier><identifier>DOI: 10.1093/eurheartj/ehz747.0001</identifier><language>eng</language><publisher>Oxford University Press</publisher><ispartof>European heart journal, 2019-10, Vol.40 (Supplement_1)</ispartof><rights>Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2019. For permissions, please email: journals.permissions@oup.com. 2019</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c1391-c1cefe2e4cfc2d4a83cf8517a2c9d2e22b8e4b91b8890761654f4bbe5efc27a13</citedby></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,777,781,1579,27905,27906</link.rule.ids></links><search><creatorcontrib>Hu, L</creatorcontrib><creatorcontrib>Sharir, T</creatorcontrib><creatorcontrib>Fish, M B</creatorcontrib><creatorcontrib>Ruddy, T D</creatorcontrib><creatorcontrib>Di Carli, M</creatorcontrib><creatorcontrib>Dorbala, S</creatorcontrib><creatorcontrib>Einstein, A J</creatorcontrib><creatorcontrib>Betancur, J</creatorcontrib><creatorcontrib>Eisenberg, E</creatorcontrib><creatorcontrib>Commandeur, F</creatorcontrib><creatorcontrib>Germano, G</creatorcontrib><creatorcontrib>Damini, D</creatorcontrib><creatorcontrib>Berman, D</creatorcontrib><creatorcontrib>Slomka, P J</creatorcontrib><title>29Prognostic safety of automatic cancellation of rest myocardial perfusion scan by machine learning: a report from multicenter REFINE SPECT registry of new generation SPECT</title><title>European heart journal</title><description>Abstract
Background
We aimed to develop a machine learning (ML) computer score derived from stress imaging and clinical data, which indicates if the rest scan could be automatically and safely canceled in the routine stress/rest myocardial perfusion SPECT (MPS).
Methods
A total of 20414 stress/rest cases from the REFINE SPECT registry collected from 5 sites in 3 countries with Tc-99m-based MPS images, clinical data, and clinical follow-up were included in the study. All images were automatically processed at our Medical Center. The automatically generated myocardial contours were checked by experienced technologists. In total, 93 variables (26 clinical, 17 stress-test, and 50 stress-imaging variables) were used to build a LogitBoost model for prediction of adverse events (AE), including coronary revascularization, death, myocardial infarction, and unstable angina. 10-fold cross-validation was performed to separate test from validation data for the assessment of ML. The overall ML predictive performance was compared to quantitative (stress total perfusion deficit [TPD]) by the area under the receiver operating characteristic curves (AUC). ML cut-off (ML1) to simulate the decision of cancellation of the rest scan was set to result in the same % of normal scans as these determined by the normal clinical reader diagnosis on a 4-point scale in the whole population, or the same % of scans with visual summed stress scores (SSS) = 0 in the subpopulation with available SSS. A second ML cutoff (ML2) was established to achieve a 1% annual risk of AE. The annual risk of AE of the normal ML score was compared with normal clinical diagnosis and with the finding of SSS = 0.
Results
The mean follow-up interval was 4.7±1.5 years. Overall, 3542 AE were observed (3.7% annual risk). The AUC for AE was higher for ML (0.780±0.005) than for stress TPD (0.698±0.006) (p<0.001). Normal clinical diagnosis was reported in 60% cases. In 70% (14242 scans) with available segmental scores, 53% had SSS=0. ML1 and ML2 thresholds were compared with normal visual diagnosis and with SSS = 0 for AE (Figure). ML1 achieved a lower annual risk (1.5%) than normal clinical diagnosis (2.1%) or SSS = 0 (1.6% versus 2.3%) (p<0.001). The more conservative ML2 threshold with a 1% annual risk of AE resulted in a 40% canceling rate.
Figure 1
Conclusion
ML could be used to automatically cancel the rest MPS scan with the same proportion as using normal visual MPS reading, but with significantly lower AE rate in stress-only scans.
Acknowledgement/Funding
R01HL089765 from the National Heart, Lung, and Blood Institute/National Institutes of Health (NHLBI/NIH)</description><issn>0195-668X</issn><issn>1522-9645</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2019</creationdate><recordtype>article</recordtype><recordid>eNqNkN9KwzAUxoMoOKePIOQFuiXpf-9kbDoYKjrBu5JmJ11Hm5SkRerT-AK-hE9muorX3pwccr7vO4cfQteUzChJ_Tl0Zg_ctIc57D_iIJ4RQugJmtCQMS-NgvAUTQhNQy-KkrdzdGHtwSmSiEYT9MXSJ6MLpW1bCmy5hLbHWmLetbrmw5_gSkBVuV6rYWLAtrjuteBmV_IKN2BkZ4ehdVKc97jmYl8qwJU7SpWquMH8-9NAo02LpdE1rrvKJYNqweDn5Wr9sMQvT8vF1mUXpW3N8QIF77gABWbcfBRcojPJKwtXv-8Uva6W28W9t3m8Wy9uN56gfkpdFSCBQSCkYLuAJ76QSUhjzkS6Y8BYnkCQpzRPkpTEjkMYyCDPIQSnjzn1pygcc4XR1hqQWWPKmps-oyQbmGd_zLOReTYwdz4y-nTX_NPyA-fwjtU</recordid><startdate>20191001</startdate><enddate>20191001</enddate><creator>Hu, L</creator><creator>Sharir, T</creator><creator>Fish, M B</creator><creator>Ruddy, T D</creator><creator>Di Carli, M</creator><creator>Dorbala, S</creator><creator>Einstein, A J</creator><creator>Betancur, J</creator><creator>Eisenberg, E</creator><creator>Commandeur, F</creator><creator>Germano, G</creator><creator>Damini, D</creator><creator>Berman, D</creator><creator>Slomka, P J</creator><general>Oxford University Press</general><scope>AAYXX</scope><scope>CITATION</scope></search><sort><creationdate>20191001</creationdate><title>29Prognostic safety of automatic cancellation of rest myocardial perfusion scan by machine learning: a report from multicenter REFINE SPECT registry of new generation SPECT</title><author>Hu, L ; Sharir, T ; Fish, M B ; Ruddy, T D ; Di Carli, M ; Dorbala, S ; Einstein, A J ; Betancur, J ; Eisenberg, E ; Commandeur, F ; Germano, G ; Damini, D ; Berman, D ; Slomka, P J</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c1391-c1cefe2e4cfc2d4a83cf8517a2c9d2e22b8e4b91b8890761654f4bbe5efc27a13</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2019</creationdate><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Hu, L</creatorcontrib><creatorcontrib>Sharir, T</creatorcontrib><creatorcontrib>Fish, M B</creatorcontrib><creatorcontrib>Ruddy, T D</creatorcontrib><creatorcontrib>Di Carli, M</creatorcontrib><creatorcontrib>Dorbala, S</creatorcontrib><creatorcontrib>Einstein, A J</creatorcontrib><creatorcontrib>Betancur, J</creatorcontrib><creatorcontrib>Eisenberg, E</creatorcontrib><creatorcontrib>Commandeur, F</creatorcontrib><creatorcontrib>Germano, G</creatorcontrib><creatorcontrib>Damini, D</creatorcontrib><creatorcontrib>Berman, D</creatorcontrib><creatorcontrib>Slomka, P J</creatorcontrib><collection>CrossRef</collection><jtitle>European heart journal</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Hu, L</au><au>Sharir, T</au><au>Fish, M B</au><au>Ruddy, T D</au><au>Di Carli, M</au><au>Dorbala, S</au><au>Einstein, A J</au><au>Betancur, J</au><au>Eisenberg, E</au><au>Commandeur, F</au><au>Germano, G</au><au>Damini, D</au><au>Berman, D</au><au>Slomka, P J</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>29Prognostic safety of automatic cancellation of rest myocardial perfusion scan by machine learning: a report from multicenter REFINE SPECT registry of new generation SPECT</atitle><jtitle>European heart journal</jtitle><date>2019-10-01</date><risdate>2019</risdate><volume>40</volume><issue>Supplement_1</issue><issn>0195-668X</issn><eissn>1522-9645</eissn><abstract>Abstract
Background
We aimed to develop a machine learning (ML) computer score derived from stress imaging and clinical data, which indicates if the rest scan could be automatically and safely canceled in the routine stress/rest myocardial perfusion SPECT (MPS).
Methods
A total of 20414 stress/rest cases from the REFINE SPECT registry collected from 5 sites in 3 countries with Tc-99m-based MPS images, clinical data, and clinical follow-up were included in the study. All images were automatically processed at our Medical Center. The automatically generated myocardial contours were checked by experienced technologists. In total, 93 variables (26 clinical, 17 stress-test, and 50 stress-imaging variables) were used to build a LogitBoost model for prediction of adverse events (AE), including coronary revascularization, death, myocardial infarction, and unstable angina. 10-fold cross-validation was performed to separate test from validation data for the assessment of ML. The overall ML predictive performance was compared to quantitative (stress total perfusion deficit [TPD]) by the area under the receiver operating characteristic curves (AUC). ML cut-off (ML1) to simulate the decision of cancellation of the rest scan was set to result in the same % of normal scans as these determined by the normal clinical reader diagnosis on a 4-point scale in the whole population, or the same % of scans with visual summed stress scores (SSS) = 0 in the subpopulation with available SSS. A second ML cutoff (ML2) was established to achieve a 1% annual risk of AE. The annual risk of AE of the normal ML score was compared with normal clinical diagnosis and with the finding of SSS = 0.
Results
The mean follow-up interval was 4.7±1.5 years. Overall, 3542 AE were observed (3.7% annual risk). The AUC for AE was higher for ML (0.780±0.005) than for stress TPD (0.698±0.006) (p<0.001). Normal clinical diagnosis was reported in 60% cases. In 70% (14242 scans) with available segmental scores, 53% had SSS=0. ML1 and ML2 thresholds were compared with normal visual diagnosis and with SSS = 0 for AE (Figure). ML1 achieved a lower annual risk (1.5%) than normal clinical diagnosis (2.1%) or SSS = 0 (1.6% versus 2.3%) (p<0.001). The more conservative ML2 threshold with a 1% annual risk of AE resulted in a 40% canceling rate.
Figure 1
Conclusion
ML could be used to automatically cancel the rest MPS scan with the same proportion as using normal visual MPS reading, but with significantly lower AE rate in stress-only scans.
Acknowledgement/Funding
R01HL089765 from the National Heart, Lung, and Blood Institute/National Institutes of Health (NHLBI/NIH)</abstract><pub>Oxford University Press</pub><doi>10.1093/eurheartj/ehz747.0001</doi></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 0195-668X |
| ispartof | European heart journal, 2019-10, Vol.40 (Supplement_1) |
| issn | 0195-668X 1522-9645 |
| language | eng |
| recordid | cdi_crossref_primary_10_1093_eurheartj_ehz747_0001 |
| source | Elektronische Zeitschriftenbibliothek - Frei zugängliche E-Journals; Oxford University Press Journals All Titles (1996-Current); Alma/SFX Local Collection |
| title | 29Prognostic safety of automatic cancellation of rest myocardial perfusion scan by machine learning: a report from multicenter REFINE SPECT registry of new generation SPECT |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-19T11%3A24%3A07IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-oup_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=29Prognostic%20safety%20of%20automatic%20cancellation%20of%20rest%20myocardial%20perfusion%20scan%20by%20machine%20learning:%20a%C2%A0report%20from%20multicenter%20REFINE%20SPECT%20registry%20of%20new%20generation%20SPECT&rft.jtitle=European%20heart%20journal&rft.au=Hu,%20L&rft.date=2019-10-01&rft.volume=40&rft.issue=Supplement_1&rft.issn=0195-668X&rft.eissn=1522-9645&rft_id=info:doi/10.1093/eurheartj/ehz747.0001&rft_dat=%3Coup_cross%3E10.1093/eurheartj/ehz747.0001%3C/oup_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rft_oup_id=10.1093/eurheartj/ehz747.0001&rfr_iscdi=true |