Residual U-Net convolutional neural network architecture for low-dose CT denoising


Low-dose CT has received increasing attention in the recent years and is considered a promising method to reduce the risk of cancer in patients. However, the reduction of the dosage leads to quantum noise in the raw data, which is carried on in the reconstructed images. Two different multilayer convolutional neural network (CNN) architectures for the denoising of CT images are investigated. ResFCN is based on a fully-convolutional network that consists of three blocks of 5×5 convolutions flters and a ResUNet that is trained with 10 convolutional blocks that are arranged in a multi-scale fashion. Both architectures feature a residual connection of the input image to ease learning. Training images are based on realistic simulations by using the XCAT phantom. The ResUNet approach shows the most promising results with a peak signal to noise ratio of 44.00 compared to ResFCN with 41.79.

Original languageEnglish
JournalCurrent Directions in Biomedical Engineering
Issue number1
Pages (from-to)297-300
Number of pages4
Publication statusPublished - 01.09.2018


Dive into the research topics of 'Residual U-Net convolutional neural network architecture for low-dose CT denoising'. Together they form a unique fingerprint.

Cite this