A Deep Learning Method for Diagnosing Coronary Artery Disease using SPECT Images of Heart

Thanh Trung Nguyen1, , Thai Ha Nguyen2, Duc Thuan Nguyen2, Hoang Minh Dang3
1 108 Military Central Hospital, No. 1, Tran Hung Dao, Hai Ba Trung, Ha Noi, Viet Nam
2 Hanoi University of Science and Technology, No. 1, Dai Co Viet, Hai Ba Trung, Ha Noi, Viet Nam
3 Military Information Technology Institute, No. 17 Hoang Sam, Cau Giay, Ha Noi, Viet Nam

Main Article Content

Abstract

Coronary artery disease (CAD) is one of the leading causes of death in the world, especially in the middle-aged and old populations. CAD treatment costs are very high when patients are at a late stage, complicated pathologies. This study investigated the efficiency of the diagnoses of CAD by a deep learning model using polar maps and slice images derived from myocardial perfusion imaging (MPI) by single photon emission computed tomography (SPECT) cameras. Data for evaluation were collected at the Department of Nuclear Medicine, 108 Military Central Hospital. The experimental results showed that learning from MPI slice images provided a higher diagnosis accuracy than from polar map images.

Article Details

References

[1] Goff D. C., Jr., et al, 2013 ACC/AHA guideline on the assessment of cardiovascular risk: a report of the American College of Cardiology, American Heart Association Task Force on Practice Guidelines, Circulation, 129, 25 Suppl 2 (2014), 49-73.
[2] Einstein AJ, Effects of radiation exposure from cardiac imaging: how good are the data?, J Am Coll Cardiol 59 (2012), 553–565.
[3] Hongkai Wang, Zongwei Zhou, Comparison of machine learning methods for classifying mediastinal lymph node metastasis of non-small cell lung cancer from 18F-FDG PET/CT images, EJNMMI Research (2017), 7(1): 11.
[4] Ryo Nakazato, Balaji K. Tamarappoo, Xingping Kang, Arik Wolak, Faith Kitel, Sean W. Hayes, Louise E.J. Thomson, John D. Friedman, Daniel S. Berman and Piotr J. Slomka, Quantitative Upright–Supine High-Speed SPECT Myocardial Perfusion Imaging for Detection of Coronary Artery Disease: Correlation with Invasive Coronary Angiography, Journal of Nuclear Medicine (2010), 51(11): 1724–1731.
[5] Yann LeCun, Yoshua Bengio & Geoffrey Hinton, Deep learning, Nature (2015), 521, 436–444.
[6] Varun Gulshan, PhD; Lily Peng, MD, PhD; Marc Coram, PhD; et al, Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs, JAMA (2016), 316(22).
[7] Betancur, Commandeur, Motlagh, Deep Learning for Prediction of Obstructive Disease from Fast Myocardial Perfusion SPECT, JACC: Cardiovascular Imaging (2018), 11(11): 1654–1663.
[8] Peter L. Tilemeker, MD, Standardized reporting of radionuclide myocardial perfusion and function, Journal of Nuclear Cardiology (2009).
[9] Bateman TM, Dilsizian V, Beanlands RS, DePuey EG, Heller GV, Wolinsky DA, American society of nuclear cardiology position statement, Journal of Nuclear Cardiology (2016).
[10] Matthew D. Zeiler and Rob Fergus, Visualizing and Understanding Convolutional Networks, ECCV 2014: Computer Vision – ECCV (2014), 818–833.
[11] Hesse B, et al, EANM/ESC procedural guidelines for myocardial perfusion imaging in nuclear cardiology, European Journal of Nuclear Medicine and Molecular Imaging 32(7) (2005), 855–897.
[12] Holly T. A., et al, Single photon-emission computed tomography, J Nucl Cardiol 17(5) (2010), 941–973.
[13] Available at Http://www3.gehealthcare.ca/en/products/categories/nuclear_medicine/xeleris_workstations_and_applications/evolutionforcardiac.
[14] Karen Simonyan, Andrew Zisserman, Very Deep Convolutional Networks for Large-Scale Image Recognition, ICLR (2015).