CT image reconstruction method and system based on convolutional neural network

Abstract

The invention discloses a CT image reconstruction method and system based on a convolutional neural network, and the method comprises the steps: inputting projection data, carrying out the filtering processing and back projection processing of the projection data through a front-end convolutional neural network, generating tomographic image information, and processing the generated tomographic image information through a back-end convolutional neural network. And calculating the value of the whole network loss function by using the processed tomographic image and the real image, reversely feeding back the gradient information of the loss function layer by layer, and updating the parameter value of each network layer. According to the invention, the projection data is processed by using the multi-layer convolutional neural network, useful information in the projection data is fully mined, and the feature extraction and information expression capabilities of the overall neural network model are effectively improved; a full connection layer is prevented from being used in the neural network model, and the required neural network model parameters are few and easy to implement.

Description

technical field [0001] The present invention relates to the field of medical technology, in particular to a convolutional neural network-based CT image reconstruction method and system. Background technique [0002] Due to the great advantages of X-ray computed tomography (CT) technology in disease screening and diagnosis, this technology has become an indispensable imaging method in the clinical practice of modern hospitals, usually including whole-body diagnostic CT and C-arm CT interventional imaging. , Dental CT, etc. There are two main factors affecting the wide application of CT technology in practice: reconstructed image quality and image reconstruction speed. In recent years, with the development of deep learning technology, the quality of CT reconstruction images has been significantly improved; with the development of GPU hardware acceleration technology, the time required for CT image reconstruction has been significantly shortened. [0003] Initially, deep lear...

AI Technical Summary

Problems solved by technology

However, only two-layer shallow convolutional neural network is used in this method, and projection data preprocessing includes projection image denoising and Ramp filtering, and the representation ability of only two-layer shallow convolutional neural network is far Far from meeting the needs of practical applications, it limits the application potential of neural networks to a certain extent, and cannot effectively mine useful information in projection data.

Images