Research Progress of Machine Learning and Deep Learning in Intelligent Diagnosis of the Coronary Atherosclerotic Heart Disease

Table 2

AI applications in CCTA.

Method	Tasks	Data	Measure	Value	Calculate time or cost	Paper
ML	Modifications to the reconstructed coronary tree	122	Average quality score		Within 2 minutes	[24]
	Identification of the degree of coronary stenosis	42	AUC	0.94	Less than 1 second	[25]
	Characterization of coronary plaques	32	DSC	83.2%	—	[26]
DL	Coronary plaque characterization and detection of coronary stenosis	163	Accuracy	77%	—	[27]
	Calculation of coronary functional parameters	1052	AUC	0.78	A few seconds	[28]
	Segmentation of left ventricular myocardium and calculation of coronary functional parameters	126	AUC		—	[30]
	Segmentation of left ventricular myocardium and calculation of coronary functional parameters	126	AUC	0.76	—	[31]