Respiratory movement prediction method

Abstract

The invention provides a respiratory movement prediction method which comprises the following steps of constructing a BP neural network, and optimizing the BP neural network by fusing a simulated annealing genetic algorithm, optimizing a weighting error of the BP neural network, acquiring respiratory movement data and processing the respiratory movement data, training the BP neural network with the optimized weighting error by using the processed respiratory movement data to obtain a strong regression device, and predicting the respiratory movement state of the human body by using a strong regression device. The respiratory movement state of the patient can be predicted with high precision, and the actual tumor is prevented from exceeding the target area, so that the patient can be better subjected to radioactive therapy.

Description

technical field [0001] The present invention relates to the field of medical technology, in particular, to a respiratory motion prediction method. Background technique [0002] During radiation therapy for tumors, the effects of breathing motion can cause changes in the position and shape of tissues and organs in the chest and abdomen. In order to prevent the actual tumor from exceeding the target area, the methods usually adopted are: 1) the method of expanding the range of radiotherapy, which will increase the radiation dose to the patient due to the expansion of the target area; 2) breath-holding technique, which requires Patients hold their breath, but many patients cannot hold their breath for a long time; 3) image acquisition method, this method will cause artifacts in the acquired image, which limits its practical utility; 4) respiratory gating technology, this method only involves Acquisition / use of imaging data during a limited window (eg end-tidal) based on simple...

AI Technical Summary

Problems solved by technology

[0004] Based on this, in order to solve the problem that the prior art cannot predict the respiratory movement state of the patient, the present invention provides a respiratory movement prediction method, and its specific technical solution is as follows:

Images