Table 1.

Application of artificial intelligence to reduce the radiation dose of CCTA.

Study	Year	Networks	Algorithm	ED (mSv)	Degree of radiation dose reduction (%)
Wolterink et al. (18)	2017	CNN	Discriminator CNN	0.2	NA
Kang et al. (19)	2018	GAN	Cycle-consistent adversarial denoising network	NA	NA
Benz et al. (20)	2022	CNN	DLIR	0.8	43
Liu et al. (21)	2020	GAN	GAN, Adversarial CNN combined with CNN	0.91	55.65
Li et al. (22)	2022	DNN	DLIR-H	0.75 ± 0.14	54.5
Sun et al. (23)	2022	DNN	DLIR	0.57 ± 0.31	36

DNN, deep neural network; GAN, generative adversary networks; CNN, convolutional neural network; CCTA, coronary computed tomography angiography; DL, deep learning; ED, effective dose; DLIR, deep learning image reconstruction; DLIR-H, high-strength deep learning image reconstruction.