Patient chest-and-belly tumor breathing movement predicting and tracking method

Abstract

The invention provides a patient chest-and-belly tumor breathing movement predicting and tracking method which comprises the following steps of continuously acquiring a body surface breathing movementby means of optical position tracking equipment; discontinuously acquiring the position of an internal tumor by means of X-ray three-dimensional planar imaging positioning equipment; performing datapreprocessing, namely performing wavelet decomposition on the body surface breathing movement, thereby obtaining a baseline, a low-frequency part which represents a main breathing signal, and a high-frequency part which represents noise; obtaining a body surface breathing movement predication result through a breathing prediction algorithm; using the acquired internal tumor position and the body surface breathing movement at a corresponding time point as an input, establishing a body surface-internal breathing movement correlation model by means of a linear and nonlinear hybrid polynomial model; reproducing the body surface-internal breathing movement correlation model, using the predicted continuous body surface breathing movement as an input, and calculating a continuous internal tumor position; and entering the correlation model and updating a discriminating logic. The method can be used for guiding execution of an accurate tracking radiotherapy.

Description

technical field [0001] The invention belongs to the field of medical signal processing, and in particular relates to a method for predicting and tracking the respiratory movement of a patient's chest and abdomen tumors, which is applied to the radiotherapy process. Background technique [0002] Lung tumors and other thoracic and abdominal tumors, which have the highest incidence rate among malignant tumors, will undergo intraoperative displacement caused by the patient's breathing. With the development of technology, the combination of X-ray stereoscopic imaging and optical real-time tracking technology makes it possible to track the breathing movement of tumors in patients in real time; and the high-precision breathing prediction algorithm can compensate for data processing and equipment electrical Time delays introduced by mechanical actions, etc. Therefore, the radiotherapy equipment can track and compensate the moving tumor during the radiotherapy process, implement mor...

AI Technical Summary

Problems solved by technology

[0005] The verification and update logic of the motion correlation model adopted by the respiratory synchronization tracking system has a hysteresis in response to changes in the respiratory model, and for a smooth breathing process, there is no basis to further reduce the number of verification images and reduce the dose received by the patient

Moreover, the hybrid prediction algorithm adopted by it has a large error in predicting irregular or abnormal breathing, and there is still a lot of room for improvement.

[0006] For other researched breathing prediction algorithms, the NN algorithm is prone to local minimization or over-fitting problems

The SVM algorithm is highly complex, and the prediction results depend on parameter selection

The ANFIS algorithm uses the position of respiratory movement as the input parameter of the model to construct a fuzzy set for prediction, which has a strong prediction ability, but the prediction error is too large for irregular signals, especially when the amplitude of the respiratory signal changes suddenly.

Images