Supine position and prone position mammary gland image registration method based on deep learning

Abstract

The invention discloses a supine position and prone position mammary gland image registration method based on deep learning. The method comprises the steps that a deep learning registration network is constructed, the registration network comprises an affine registration network, a first spatial transformation network, an elastic registration network and a second spatial transformation network, and in an up-sampling structure of the elastic registration network, each up-sampling layer outputs a deformation field; and the registration network is trained, a deformation field and a loss function value between a fixed image and a transformed moving image are calculated until a set total loss function meets an optimization convergence condition, the fixed image is a supine position or prone position mammary gland image, and the moving image is a prone position or supine position mammary gland image with different body positions from the fixed image. The method has the advantages of being high in registration speed, simple in model, high in generalization performance and capable of generating less deformation which does not conform to reality.

Description

technical field [0001] The present invention relates to the technical field of medical image processing, and more specifically, to a breast image registration method for supine and prone positions based on deep learning. Background technique [0002] The purpose of image registration is to obtain one or a series of transformations between two images, so that the corresponding points in the two images achieve the same spatial position. These two images are called a fixed image and a moving image, respectively. Supine and prone breast images refer to the positions in which the patient is in the supine and prone positions when the medical images are taken. Since the breast tissue is a soft tissue and the patient's position changes, the shape of the breast changes greatly in different positions, which makes it more difficult to register breast images in the supine and prone positions. Breast image registration in the supine and prone positions has potential applications in bre...

AI Technical Summary

Problems solved by technology

Therefore, in the existing breast configuration method, the number of networks is large, the amount of calculation and parameters are large, and due to the limited training data set, overfitting is prone to occur, and the generated deformation field has many inconsistencies. actual deformation

Images