Tumor clinical target volume invasiveness probability calculating method based on hidden Markov model

Abstract

The invention discloses a method for calculating the invasion probability of a tumor clinical target area based on a hidden Markov model. The frequency of joint appearance of grid points in the three-dimensional tumor area is used to obtain the corresponding correlation rules; the primary tumor data of the new patient is read in, and the distance transformation is used to obtain the spatial distance from each voxel point to the primary tumor area; The probability of the primary tumor area is set as 1, and the invasion probability of each voxel point is related to the probability of adjacent voxels and the probability of the last observed state, and iteratively calculates outward from the primary tumor area; finally, all voxels are calculated The probability of a point being invaded by a tumor. The invention can assist the formulation of clinical radiotherapy plans, realize the automatic determination of the range of clinical radiotherapy target areas with different emission doses, reach the delineation standard for clinical application, and improve the working efficiency of doctors.

Description

technical field [0001] The invention relates to the field of predicting tumor growth and simulating tumor spread, in particular to a method for calculating the invasion probability of a tumor clinical target area based on a hidden Markov model. Background technique [0002] Radiation therapy is the mainstream treatment method for treating tumors at present, and it is suitable for the treatment of various tumors, and the precise determination of the range of radiotherapy target area is the key factor for the success of radiotherapy. Accurate radiotherapy target area can effectively reduce the probability of tumor recurrence and reduce the side effects of radiotherapy. During the specific treatment process, doctors will determine different radiotherapy target areas and radiation doses according to the different probabilities of different tissues being invaded by tumors. At present, the determination of radiotherapy target volume mainly depends on the doctor's clinical experie...

AI Technical Summary

Problems solved by technology

[0003] Due to the existence of objective reasons (only the perspective reference of the horizontal plane and the lack of reference of the upper and lower planes of view) and subjective reasons (subjective differences caused by different experience of different radiologists), the target area drawn manually at present results in different target volumes between different doctors. The delineation results of the districts vary widely

The determination of the follow-up treatment plan and the evaluation of the treatment effect have brought a negative impact

On the other hand, it usually takes 2-3 hours to delineate a patient's radiotherapy target area, which requires a lot of manpower and time costs, and the delineation system of ABAS is too simple, and a single distance extension is not suitable for complex structures and tissues.

Images