An image synthesis method based on a convolution neural network

Abstract

The invention discloses an image synthesis method based on a convolution neural network, comprising the following steps: Step 1: Pre-processing the 3T-7T training image pair and the test 3T magnetic resonance image to obtain the pre-processed 3T-7T training image pair and the pre-tested 3T magnetic resonance image; Step 2 Constructing a dual-domain convolutional neural network model, inputting thepre-processed 3T-7T training image the dual-domain convolutional neural network model for training, and obtaining the trained dual-domain convolutional neural network model; Step 3, using processed test 3T magnetic resonance image as the current 3T magnetic resonance image, and inputing the current image into the trained dual domain convolutional neural network model to obtain a 7T magnetic resonance image corresponding to the current 3T magnetic resonance image.

Description

technical field [0001] The invention belongs to the field of image processing, and in particular relates to an image synthesis method based on a convolutional neural network. Background technique [0002] With the rapid development of magnetic resonance imaging technology, the resolution, signal-to-noise ratio and scanning speed of magnetic resonance images have been greatly improved. However, the current 7T MRI scanners are extremely expensive, and the distribution is very scarce, with less than 100 in the world. In contrast, as a routine clinical choice, 3T MRI scanners have become the gold standard in the industry since the early 20th century, and are still commonly used in scientific research and clinical practice today. In order to improve the quality of magnetic resonance images, the common clinical measures are to use smaller voxels for image acquisition, which can obtain more image details, higher resolution and contrast, but at the same time cause low signal-to-noi...

AI Technical Summary

Problems solved by technology

[0005] In view of the deficiencies in the prior art, the purpose of the present invention is to provide an image synthesis method based on a convolutional neural network to solve the technical problem that the prior art cannot efficiently utilize 3T magnetic resonance images to obtain 7T magnetic resonance images

Images