Machine learning for diagnosis of coronary artery disease in computed tomography angiography: A survey

Table 1 Summary of different machine learning-based methods used in coronary artery disease diagnosis

CAD diagnosis	Method	Task	Category
Coronary artery extraction
Schaap et al[14]	Linear and nonlinear regression	Artery	ML
Huang et al[15]	3D U-net	Artery	DL
Kong et al[16]	ConvRNN + ConvGRU	Artery	DL
Shen et al[17]	3D FCN + level set	Artery	DL
Wu et al[18]	CNN + nearest neighbor search	Artery	DL
Wolterink et al[19]	3D dilated CNN	Centerline	DL
Coronary plaque detection
Mittal et al[20]	PBT, RF	Calcified	ML
Kurkure et al[21]	SVM	Calcified	ML
Wei et al[22]	Linear discriminant analysis	Soft	ML
Jawaid et al[23]	SVM	Soft	ML
Tessmann et al[24]	AdaBoost	Multiple	ML
Kelm et al[25]	PBT, RF	Multiple	ML
Zhao et al[26]	SVM	Multiple	ML
Zreik et al[27]	CNN + RNN	Multiple	DL
Huo et al[28]	Attention recognition dual network	Calcified	DL
Vulnerable plaque identification
Kolossváry et al[33]	Radiomics	NRS	ML
Kolossváry et al[2]	Radiomics	LAP &NRS	ML
Kolossváry et al[34]	Logistic regression, K-nearest neighbors, RF, least angle regression, naive Bayes, Gaussian process classifier, decision trees, DNN	Advanced lesion	ML, DL
Coronary stenosis assessment
Zuluaga et al[36]	SVM	ASS	ML
Kang et al[37]	SVM + formula-based analytical method	ASS	ML
Zreik et al[27]	CNN + RNN	ASS	DL
Itu et al[41]	DNN	HSS	DL
Wang et al[42]	DeepVessel-FFR	HSS	DL
Dey et al[43]	Boosted ensemble algorithm	HSS	ML
Kumamaru et al[44]	2D conditional generative adversarial network + 3D convolutional ladder network	HSS	DL

ASS: Anatomically significant stenosis; CAD: Coronary artery disease; CNN: Convolutional neural network; ConvGRU: Convolutional gated recurrent unit; ConvRNN: Convolutional recurrent neural network; DL: Deep learning method; DNN: Deep neural network; FCN: Fully convolutional network; FFR: Fractional flow reserve; HSS: Hemodynamically significant stenosis; LAP: Low density plaque; ML: Conventional machine learning method; NRS: Napkin ring sign; PBT: Probability boosting tree; RF: Random forest; RNN: Recurrent neural network; SVM: Support vector machine.