Study Of The Methods For Improving The Quality And Efficiency Of Intensity Modulated Radiotherapy Plan

The accurately and efficiently planning in the radiotherapy treatment planning system plays an important role in the implementation of a precise radiotherapy treatment,which highly depends on the dose calculation algorithms,optimization models,optimization algorithms of solving reverse problem as well as the intelligence degree of planning process.In order to improve the quality of radiotherapy plan,this paper makes an in–depth study on the radiotherapy optimization model,and the choice of the initial beam weights for dose–volume–based optimization（DVO）based on the gradient optimization algorithm.In order to improve the efficiency of radiotherapy plan,this paper studies the automatic optimization algorithm of radiotherapy parameters.The main innovative work of this paper are summarized as follows:1.According to the theory of Pareto surface generated by the organ–based optimization model belonging to the Pareto surface generated by the dose–volume–based optimization,we proposed a method of determing the initial beam weights for dose–volume–based optimization based on the gradient optimization algorithm.To improve the quality of DVO radiotherapy plan and reduce the local minimum error generated by non-convexity of the dose–volume objectives,we proposed an efficient method（an organ–based optimization model guiding DVO）to determine the initial beam weights for DVO.First,fluence map optimization applying improved organ–based optimization model was performed.Then,DVO was performed by using the beam weights determined in the first step as its initial beam weights.We demonstrated this technique in five prostate cancer cases and five head–and–neck cancer cases.The experimental results show that compared with the conventional DVO plan with uniform initial beam weights,the improved DVO plan provided better protection of organs at risk,while the planning target volume（PTV）coverage was similar.Moreover,the improved DVO plan was better than the plan generated in the first step.The proposed method,with advantages in determining the initial beam weights,is highly efficient to improve DVO plans and reduce the DVO optimization time.2.The improved functions of equivalent uniform dose（EUD）linear sub–score,maximum dose sub–score and quadratic EUD sub–score are proposed.（1）The EUD linear sub–score limits the optimization ability of optimization algorithm,because it does not differentiate the doses that satisfy the prescription doses,gradient vanish and derivatives discontinuous in the feasible solution space.To solve the problem,this paper based on regularization theory,presents a smooth and convex regularization EUD sub–score,named lse（log–sum–exp）EUD sub–score.Then,a hybrid radiotherapy optimization model using lse EUD sub–score was constructed,and the validity of the new model was verified on four kinds of testing cases.Compared to the original model using EUD linear sub-score,experimental results show that the method using newly proposed model using lse EUD sub–score can get better protection to organs at risks,while ensuring similar dose coverage for target,the quality of radipotherapy plan was improved.（2）The maximum–dose–based and generalized equivalent uniform dose（gEUD）–based quadratic sub–scores limit the optimization ability of optimization algorithm,because they do not differentiate the doses that satisfy the prescription doses,gradient vanish and derivatives discontinuous in the feasible solution space.To address these drawbacks,this study proposes new sub-scores for the maximum dose criterion and the gEUD criterion based on the theory of piecewise penalty.In the new piecewise sub–scores,doses lower than the prescribed dose are assigned a linear penalty function,and those higher than the prescribed dose are assigned an extra quadratic penalty function.To test their efficiency,they were incorporated into a physical model and a hybrid physical–biological model,respectively,and were tested on four kinds of testing cases.For similar or better PTV coverage,optimization based on our proposed quadratic models is capable of improving the OARs sparing.In addition,compared with the optimization based on DV constraints,the new optimization models have less number of parameters to be adjusted,so the complexity of planning is reduced.Our proposed optimization method has the potential to expand the search ability in the the feasible solution space and improve the quality of radiotherapy plan.3.After understanding the existing automatic optimization techniques in radiotherapy,two automatic optimization methods of radiotherapy parameters were proposed.（1）To improve the efficiency of radiotherapy planning and reduce the manual intervention,an automatic method for the optimization of importance factors based on prescribed dose was proposed.First,the importance factors are automatically and iteratively adjusted based on our proposed penalty strategies.Then,Plan evaluation is performed to determine whether the obtained plan is acceptable.If not,a higher penalty is assigned to the unsatisfied objective by multiplying it by a compensation coefficient.The optimization processes are performed alternately until an acceptable plan is obtained or the maximum iteration N_max of the step three is reached.Tested on 10 cases of prostate cancer and compared with manual method,the results showed that the quality of the proposed automatic plan was comparable to,or even better than,the manual plan in terms of the dose volume histogram（DVH）and dose distributions.The proposed algorithm has potential to significantly improve the efficiency of the existing manual adjustment methods for importance factors,and contributes to the development of fully automated planning.Especially,the more the sub–objective functions,the more obvious the advantage of our algorithm.（2）In order to incorporate the parameter adjusting experiences of physicist into the automatic optimization process,a parameter automatic optimization method based on fuzzy reasoning for hybrid radiotherapy was proposed.First,initialize the parameters of objective function,then automatically and iteratively adjuste the prescribed dose or importance factors applying the automatic parameter optimization system based on fuzzy inference,and optimize the fluence map parameters using the fluence map optimization module,finally the parameters of prescription dose and importance factors are determined and the acceptable solution is obtained.When the acceptable plan is generated,the algorithm tries to further improve the plan.The effectiveness of the new automatic optimization algorithm was tested in 10 cases of prostate cancer.The experimental results show that the quality of the proposed automatic plan was comparable to,or even better than,the manual plan.