Table 1. Quantitative evaluation of the different methods for the different slices in Figure 7.

Methods	Slice one				Slice two				Slice three
	PSNR	SSIM	rRMSE		PSNR	SSIM	rRMSE		PSNR	SSIM	rRMSE
EM + Gaussian filter	31.185	0.896	0.349		29	0.899	0.552		35.561	0.95	0.477
EM + NLM	34.883	0.926	0.228		36	0.929	0.268		39.737	0.96	0.295
CNN	35.86	0.937	0.204		36	0.949	0.231		38.085	0.952	0.357
GAN	36.048	0.939	0.196		34	0.934	0.324		37.074	0.928	0.401
LCPR-Net	36.864	0.947	0.181		36	0.958	0.223		41.823	0.966	0.232

PSNR, peak signal-to-noise ratio; SSIM, structural similarity index; rRMSE, relative root mean square error; EM, expectation-maximization; NLM, nonlocal means; CNN, convolutional neural network; GAN, generative adversarial network; LCPR-Net, low-count positron emission tomography image reconstruction network.