TRAINING METHOD AND SYSTEM FOR LOW-DOSE CT IMAGE DENOISING NETWORK

摘要

The present invention provides a training method for a low-dose CT image denoising network, comprising the steps of: obtaining a training data set; establishing a low-dose CT image denoising network, which comprises a first convolutional layer, a convolution module, a first fusion layer and a second convolutional layer which are connected in sequence, the first fusion layer being used for fusing an input signal of the convolution module and an output signal of the convolution module, the convolution module comprising at least one convolutional network which is connected in sequence, each convolutional network comprising a channel layer, a third convolutional layer and a second fusion layer which are connected in sequence, the channel layer comprising a first channel which comprises a fourth convolutional layer and a first deconvolutional layer, and the fourth convolutional layer and the first deconvolutional layer being alternately connected; and training the low-dose CT image denoising network by using the training data set. According to the training method for the low-dose CT image denoising network of the present invention, by alternately connecting the fourth convolutional layer and the first deconvolutional layer, information loss can be avoided.