| With the increasing incidence and mortality of cancer around the world,it is predicted that cancer will surpass cardiovascular and cerebrovascular diseases in the visible future,and become the number one killer that threatens human health[1].Radiation therapy is one of the three major treatments for cancer and has a high clinical usage rate.However,the current dose accuracy of radiation therapy largely restricts the clinical cure rate and the quality of life.Therefore,in the course of radiotherapy,we can verify and improve the radiotherapy plan in time by reconstructing the spatial dose distribution in the target area of the patient,which is of great significance to the improvement of the accuracy and effect of radiotherapy.At present,the traditional Monte Carlo algorithm has lost its clinical utility temporarily as its running speed is too slow.But on the other hand it is undeniable that Monte Carlo is currently the most accurate method for calculating the dose of radiotherapy.This article focuses on the core issue of dose guided radiation therapy after extensive and systematic literature research and do a systematic study on the key algorithms or techniques related to3D dose reconstruction and validation by using Monte Carlo as the principle basis and verification standard.Finally the implementation of the core algorithm in dose guided radiation therapy is basically completed.This paper,based on the calculation of three dimensional dose and the error detection of radiation therapy,puts forward the corresponding solution algorithm and verifies it by the experiment.In the aspect of three-dimensional dose reconstruction,we expand and improve the related algorithms in CT reconstruction technology to form a back-projection dose reconstruction algorithm,which complete the whole algorithm flow from data acquisition by electronic field image device(EPID)to dose reconstruction and gain a good result.In the aspect of position error and target area error detection in radiotherapy,this paper propose an algorithm of dimensionality reduction registration,which solved the problem of poor registration effect between KV projection and low pixel contrast MV projection images.Finally,we verify the effectiveness and feasibility of the algorithm by the simulation of the MATLAB program,which provides a certain algorithm support for the clinical application of intensity modulated radiation therapy(IMRT). |