Literature DB >> 23482666

Improved accuracy of myocardial perfusion SPECT for the detection of coronary artery disease using a support vector machine algorithm.

Reza Arsanjani1, Yuan Xu, Damini Dey, Matthews Fish, Sharmila Dorbala, Sean Hayes, Daniel Berman, Guido Germano, Piotr Slomka.   

Abstract

UNLABELLED: We aimed to improve the diagnostic accuracy of automatic myocardial perfusion SPECT (MPS) interpretation analysis for the prediction of coronary artery disease (CAD) by integrating several quantitative perfusion and functional variables for noncorrected (NC) data by Support Vector Machine (SVM) algorithm, a computer method for machine learning.
METHODS: Rest-stress gated (99m)Tc MPS NC studies (n = 957) from 623 consecutive patients with correlating invasive coronary angiography and 334 with a low likelihood of CAD (<5%) were assessed. Stenosis ≥50% in left main or ≥70% in all other vessels was considered abnormal. Total perfusion deficit (TPD) was computed automatically. In addition, ischemic changes (ISCHs) and ejection fraction changes (EFCs) between stress and rest were derived by quantitative software. The SVM was trained using a group of 125 patients (25 with low-likelihood, 25 with 0-vessel, 25 with 1-vessel, 25 with 2-vessel, and 25 with 3-vessel CAD) with the above quantitative variables and second-order polynomial fitting. The remaining patients (n = 832) were categorized using probability estimates, with CAD defined as a probability estimate ≥ 0.50. The diagnostic accuracy of SVM was also compared with visual segmental scoring by 2 experienced readers.
RESULTS: The sensitivity of SVM (84%) was significantly better than ISCH (75%, P < 0.05) and EFC (31%, P < 0.05). The specificity of SVM (88%) was significantly better than TPD (78%, P < 0.05) and EFC (77%, P < 0.05). The diagnostic accuracy of SVM (86%) was significantly better than TPD (81%), ISCH (81%), or EFC (46%) (P < 0.05 for all). The receiver-operating-characteristic (ROC) area under the curve for SVM (0.92) was significantly better than TPD (0.90), ISCH (0.87), and EFC (0.64) (P < 0.001 for all). The diagnostic accuracy of SVM was comparable to the overall accuracy of both visual readers (86% vs. 84%, P = NS). The ROC area under the curve for SVM (0.92) was significantly better than that of both visual readers (0.87 and 0.88, P < 0.03).
CONCLUSION: Computational integration of quantitative perfusion and functional variables using the SVM approach significantly improves the diagnostic accuracy of MPS and can significantly outperform visual assessment based on ROC analysis.

Entities:  

Mesh:

Year:  2013        PMID: 23482666      PMCID: PMC3615055          DOI: 10.2967/jnumed.112.111542

Source DB:  PubMed          Journal:  J Nucl Med        ISSN: 0161-5505            Impact factor:   10.057


  35 in total

1.  Universal learning curves of support vector machines.

Authors:  M Opper; R Urbanczik
Journal:  Phys Rev Lett       Date:  2001-05-07       Impact factor: 9.161

2.  Radius margin bounds for support vector machines with the RBF kernel.

Authors:  Kai-Min Chung; Wei-Chun Kao; Chia-Liang Sun; Li-Lun Wang; Chih-Jen Lin
Journal:  Neural Comput       Date:  2003-11       Impact factor: 2.026

3.  On the accuracy and reproducibility of quantitative gated myocardial perfusion SPECT.

Authors:  G Germano; D S Berman
Journal:  J Nucl Med       Date:  1999-05       Impact factor: 10.057

4.  Automated quantification of myocardial perfusion SPECT using simplified normal limits.

Authors:  Piotr J Slomka; Hidetaka Nishina; Daniel S Berman; Cigdem Akincioglu; Aiden Abidov; John D Friedman; Sean W Hayes; Guido Germano
Journal:  J Nucl Cardiol       Date:  2005 Jan-Feb       Impact factor: 5.952

5.  Combined supine and prone quantitative myocardial perfusion SPECT: method development and clinical validation in patients with no known coronary artery disease.

Authors:  Hidetaka Nishina; Piotr J Slomka; Aiden Abidov; Shunichi Yoda; Cigdem Akincioglu; Xingping Kang; Ishac Cohen; Sean W Hayes; John D Friedman; Guido Germano; Daniel S Berman
Journal:  J Nucl Med       Date:  2006-01       Impact factor: 10.057

6.  Quantitation in gated perfusion SPECT imaging: the Cedars-Sinai approach.

Authors:  Guido Germano; Paul B Kavanagh; Piotr J Slomka; Serge D Van Kriekinge; Geoff Pollard; Daniel S Berman
Journal:  J Nucl Cardiol       Date:  2007-07       Impact factor: 5.952

7.  Effectiveness of nuclear cardiology training guidelines: a comparison of trainees with experienced readers.

Authors:  R J Golub; J R McClellan; S D Herman; M I Travin; G M Kline; P W Aitken; J I Baron; A W Ahlberg; G V Heller
Journal:  J Nucl Cardiol       Date:  1996 Mar-Apr       Impact factor: 5.952

8.  Interpretive reproducibility of stress Tc-99m sestamibi tomographic myocardial perfusion imaging.

Authors:  R J Golub; A W Ahlberg; J R McClellan; S D Herman; M I Travin; J F Mather; P W Aitken; J I Baron; G V Heller
Journal:  J Nucl Cardiol       Date:  1999 May-Jun       Impact factor: 5.952

9.  Identification of severe or extensive coronary artery disease in women by adenosine technetium-99m sestamibi SPECT.

Authors:  A M Amanullah; D S Berman; R Hachamovitch; H Kiat; X Kang; J D Friedman
Journal:  Am J Cardiol       Date:  1997-07-15       Impact factor: 2.778

10.  Comparative prognostic value of automatic quantitative analysis versus semiquantitative visual analysis of exercise myocardial perfusion single-photon emission computed tomography.

Authors:  D S Berman; X Kang; K F Van Train; H C Lewin; I Cohen; J Areeda; J D Friedman; G Germano; L J Shaw; R Hachamovitch
Journal:  J Am Coll Cardiol       Date:  1998-12       Impact factor: 24.094

View more
  25 in total

1.  Relationship Between Quantitative Adverse Plaque Features From Coronary Computed Tomography Angiography and Downstream Impaired Myocardial Flow Reserve by 13N-Ammonia Positron Emission Tomography: A Pilot Study.

Authors:  Damini Dey; Mariana Diaz Zamudio; Annika Schuhbaeck; Luis Eduardo Juarez Orozco; Yuka Otaki; Heidi Gransar; Debiao Li; Guido Germano; Stephan Achenbach; Daniel S Berman; Aloha Meave; Erick Alexanderson; Piotr J Slomka
Journal:  Circ Cardiovasc Imaging       Date:  2015-10       Impact factor: 7.792

2.  Fully automated analysis of perfusion data: The rise of the machines.

Authors:  Rupa M Sanghani; Rami Doukky
Journal:  J Nucl Cardiol       Date:  2017-04-21       Impact factor: 5.952

Review 3.  Cardiac imaging: working towards fully-automated machine analysis & interpretation.

Authors:  Piotr J Slomka; Damini Dey; Arkadiusz Sitek; Manish Motwani; Daniel S Berman; Guido Germano
Journal:  Expert Rev Med Devices       Date:  2017-03       Impact factor: 3.166

4.  Computer-aided diagnosis system outperforms scoring analysis in myocardial perfusion imaging.

Authors:  Lena Johansson; Lars Edenbrandt; Kenichi Nakajima; Milan Lomsky; Sven-Eric Svensson; Elin Trägårdh
Journal:  J Nucl Cardiol       Date:  2014-01-18       Impact factor: 5.952

5.  Leveraging latest computer science tools to advance nuclear cardiology.

Authors:  Piotr Slomka
Journal:  J Nucl Cardiol       Date:  2019-09-05       Impact factor: 5.952

6.  Novel SPECT Technologies and Approaches in Cardiac Imaging.

Authors:  Piotr Slomka; Guang-Uei Hung; Guido Germano; Daniel S Berman
Journal:  Cardiovasc Innov Appl       Date:  2016-12-01

7.  Machine learning for nuclear cardiology: The way forward.

Authors:  Sirish Shrestha; Partho P Sengupta
Journal:  J Nucl Cardiol       Date:  2018-04-20       Impact factor: 5.952

Review 8.  Artificial intelligence in personalized cardiovascular medicine and cardiovascular imaging.

Authors:  Ikram-Ul Haq; Iqraa Haq; Bo Xu
Journal:  Cardiovasc Diagn Ther       Date:  2021-06

9.  Prediction of cardiac death after adenosine myocardial perfusion SPECT based on machine learning.

Authors:  David Haro Alonso; Miles N Wernick; Yongyi Yang; Guido Germano; Daniel S Berman; Piotr Slomka
Journal:  J Nucl Cardiol       Date:  2018-03-14       Impact factor: 5.952

Review 10.  Automated Quantitative Nuclear Cardiology Methods.

Authors:  Manish Motwani; Daniel S Berman; Guido Germano; Piotr Slomka
Journal:  Cardiol Clin       Date:  2015-10-20       Impact factor: 2.213

View more

北京卡尤迪生物科技股份有限公司 © 2022-2023.