Fig. 3From: Quantitative evaluation of deep convolutional neural network-based image denoising for low-dose computed tomographyDenoised image with contaminating noise artifacts, which was simulated to minimize dose exposure in the image at four different dose levels. a Results of noise reduction in the entire abdominal image; b Enlarged images of the region indicated by the red region of interest; all images are presented with the same window width and level. Median, Gaussian, Wiener, original DnCNN, and optimized DnCNN (DnCNN_Tra) techniques were applied for noise reduction at each dose-reduction levelBack to article page