Artificial Intelligence in Nuclear Cardiology: Adding Value to Prognostication
- 43 Downloads
Purpose of the Review
Radionuclide myocardial perfusion imaging (MPI) continues to be an accurate and reproducible method of diagnosing obstructive coronary artery disease (CAD) with predictive, prognostic, and economic value. We review the evolutionary potential of machine learning (ML), a subset of artificial intelligence, as an adjunct to MPI.
Applying the broad scope of ML, including the integration of deep learning, can leverage the knowledge representation and automated reasoning to detect and extrapolate patterns from high-dimensional features of MPI. There is growing evidence to suggest superior abilities of ML over parametric statistical models for predicting the presence of obstructive CAD, the need for revascularization, and the occurrence of major adverse cardiac events including cardiac death.
ML is uniquely positioned to provide the next great advancement in the field of nuclear cardiology for improving patient-specific risk stratification.
KeywordsCoronary artery disease Myocardial perfusion imaging Machine learning Artificial intelligence Nuclear cardiology
Compliance with Ethical Standards
Conflict of Interest
Karthik Seetharam declares that he has no conflict of interest.
Sirish Shresthra declares that he has no conflict of interest.
James D. Mills declares that he has no conflict of interest.
Partho P. Sengupta is a consultant for HeartSciences, Ultromics, Kencor Health.
Human and Animal Rights and Informed Consent
This article does not contain any studies with human or animal subjects performed by any of the mentioned authors.
Papers of particular interest, published recently, have been highlighted as: • Of importance •• Of major importance
- 4.Shrestha S, Sengupta PP. Machine learning for nuclear cardiology: the way forward. J Nucl Cardiol. 2018.Google Scholar
- 7.Hachamovitch R, Hayes SW, Friedman JD, Cohen I, Berman DS. Stress myocardial perfusion single-photon emission computed tomography is clinically effective and cost effective in risk stratification of patients with a high likelihood of coronary artery disease (CAD) but no known CAD. J Am Coll Cardiol. 2004;43(2):200–8.CrossRefPubMedGoogle Scholar
- 15.• Haro Alonso D, Wernick MN, Yang Y, Germano G, Berman DS, Slomka P. Prediction of cardiac death after adenosine myocardial perfusion SPECT based on machine learning. J Nucl Cardiol. 2018; Alonso et al utilized a machine learning algorithm to predict cardiac death by integrating clinical and imaging data in more than 8,000 patients. The ML algorithm demonstrated higher accuracy than conventional statistical techniques. Google Scholar
- 18.•• Betancur J, Commandeur F, Motlagh M, Sharir T, Einstein AJ, Bokhari S, et al. Deep learning for Prediction of Obstructive Disease From Fast Myocardial Perfusion SPECT: a multicenter study. JACC Cardiovasc Imaging. 2018;11(11):1654–63 It is a large multi-institutional study using deep learning to predict obstructive coronary artery disease in large patient population. There are limited studies showing the role of deep learning in nuclear cardiology. CrossRefPubMedGoogle Scholar
- 19.Nakajima K, Kudo T, Nakata T, Kiso K, Kasai T, Taniguchi Y, et al. Diagnostic accuracy of an artificial neural network compared with statistical quantitation of myocardial perfusion images: a Japanese multicenter study. Eur J Nucl Med Mol Imaging. 2017;44(13):2280–9.CrossRefPubMedPubMedCentralGoogle Scholar