Table 1. Summary of related studies.

Ref.	Data type	Methods	Results
Wang et al. (2020a)	angiographic videos	adding a 3D convolutional layer to the input layer of the 2D CE–Net. (3D–2D CE–Net)	0.9855
Fan et al. (2018)	X-ray angiograms	multichannel FCN model (MSN-A)	0.9881
Fan et al. (2019)	X-ray angiography	hierarchical dense matching framework	Results for RMS errors of matches (pix): 9.79 ± 3.9
Jo et al. (2019)	X-ray angiography	selective feature mapping	Precision = 0.066, Recall = 0.091, specificity = 0.001, F1 score = 0.094 Acc = 0.980
Song et al. (2020)	X-ray angiographic	spatio-temporal constrained online layer separation (STOLS) method	the local and global rCNRs of the final vessel layer reached 2.54 and 1.24, respectively
Song et al. (2019)	X-ray angiographic image	novel inter/intra-frame constrained vascular segmentation method	Pre = 0.7378 Sen = 0.7960 F1 value = 0.7658
Wan et al. (2018)	X-ray angiography	multiscale-based adaptive Hessian-based enhancement method and a statistical region merging technique (Hessian matrix; Statistical region merging)	accuracy of 93%
Zhao et al. (2019)	X-ray angiography	automatic morphology estimation method by using directed graph construction for X-ray angiography images.	accuracy of 97.44%
Xiao, Li & Jiang (2020)	angiography CT images PASCAL VOC2012 dataset	three-dimensional U-net convolutional neural network	the dice coefficient of 0.8291
Sheng et al. (2019)	CT coronary angiography	Pre-processing: Adaptive vascular enhancement Feature extraction: automatic seed point detection Hessian matrix	0.863
Abdar et al. (2019a)	CAD datasets (Z-Alizadeh Sani and Cleveland)	novel nested ensemble nu-Support Vector Classification (NE-nu-SVC) model	accuracy of 94.66%
Yao et al. (2020)	electrocardiograph (ECG)	Gaussian naive Bayes Support vector machine XGBoost ResNet-18	h 96.16% accuracy