Hi’CT: Conceptual Design And Simulation Study Of Heavy Ion Therapy Computed Tomography Image Guide Device

Cancer is one of the main causes of death cases in the world every year,so conquer-ing cancer has become the ultimate goal pursued by all mankind.At present,the main means of treating cancer are surgery,radiotherapy and chemotherapy.In the field of ra-diotherapy,external beam therapy based on MV class X-rays dominates.Conventional X-ray radiotherapy techniques tend to deposit additional radiation doses in healthy tis-sue.Thanks to the protective effect of the Bragg peak effect on normal tissues,particle radiation therapy techniques(proton,heavy ion,etc.)have developed rapidly in the past decade.However,treatment planning for current particle radiation therapy techniques is usually still based on X-CT images.Since the interaction between X-ray photon and matter is essentially different from that of proton and heavy ion,large uncertainties will be introduced in the process of calculating the Relative Stopping Power(RSP)of the human body based on CT images.If charged particles are used for imaging,an accurate estimate of the RSP of the charged particle can be obtained directly,thereby effectively improving the accuracy of particle range calculation through RSP.In order to solve the problem of RSP accuracy in the process of carbon ion cancer treatment,combined with the development needs of Heavy Ion Research Facility in Lanzhou(HIRFL),this study designed a set of compact digital charged particle CT scanning imaging system based on Geant4 Monte Carlo simulation software——Hi’CT.In the simulation,the 430 Me V/u pencil-shaped carbon ion beam used by the Wuwei Carbon Ion Therapy System was se-lected as the source item,and its Full Width at Half Maximum(FWHM)was 3 mm.The design of the imaging device has an effective water-equivalent path length(WEPL)de-tection range of 100-250 mm,RSP resolution<1%,and can perform fine tomographic imaging of human body equivalent tissue materials.In this study,a compact digital detector is designed based on the classic architecture of”tracker-calorimeter”,and the main design indicators of the Hi’CT system are optimized by using Geant4;3 different sino space reconstruction algorithms based on tracking and non-tracking algorithms are designed,and obtain tomographic images through FBP algorithm based on GPU ac-celeration;the standard CTP404,CTP528 modules and customized CTP515 module of Catphan~?600 series phantoms are used to evaluate the imaging quality,and the specific modeling method and image quality evaluation index are given.Based on images reconstructed from data generated by Geant4 simulations,the performance of the Hi’CT system was evaluated.The core of the Hi’CT system is the detector and algorithm.The energy-range output accuracy deviation of the detector is less than 1%,which means it has good energy response performance.The performance evaluation of the image reconstruction algorithm in the sino space is mainly by compar-ing the tracking reconstruction algorithm with the ideal reconstruction algorithm at the event level.Using the=0°imaging data of the CTP404 module,it is found that after selecting an energy threshold of 10 Me V,the application of the tracking algorithm at the event level will have an accuracy of over 94%,and has a good tracks reconstruction ability.On this basis,the Hi’CT system performance is evaluated by reconstructed im-ages.For image quality,the sino space and image space images of the standard CTP404module are used for analysis.The simulation results show that the results of the tracking imaging algorithm are slightly inferior to the ideal imaging algorithm,but better than the fast imaging algorithm.The three image reconstruction algorithms all have ring and stripe artifacts.For RSP resolution,the image space images of the standard CTP404module are used for analysis.The RSP resolution of all insert materials in the tracking reconstruction algorithm and the ideal reconstruction algorithm is better than 1%.The RSP resolution of insert materials except PMP in the image reconstructed by fast recon-struction algorithm is less than 2%,and the RSP resolution of PMP is also better than2.5%.The RSP of all image reconstruction algorithms is basically consistent with the measured value.For spatial resolution,the image space images of the standard CTP528module are used for analysis.The three image reconstruction algorithms reconstructed a relatively complete line pair structure,but the image uniformity of the aluminum line pair in the image of the fast reconstruction algorithm was slightly poor.It is more se-riously affected by ring artifacts,and its MTF curve reaches the 10%point at 3 lp/cm.The image quality of the tracking reconstruction algorithm is close to that of the ideal reconstruction algorithm,and the MTF curves of these two image reconstruction algo-rithms both reach a performance of about 20%at 5 lp/cm.For head and neck imaging possibilities,image space images were analyzed using a custom CTP515 module.The three image reconstruction methods can well distinguish high-contrast tissues such as bones and soft tissues.But the fast reconstruction algorithm has more artifacts.For muscles,fats and eye lens with similar densities,the tracking reconstruction algorithm and the ideal reconstruction algorithm can distinguish them to a certain extent.Fast re-construction algorithms cannot effectively reconstruct the geometry and image contrast of specific materials.The Hi’CT system has an excellent performance in terms of detectors and algo-rithms.It has the possibility of being used for head and neck imaging,which also has the potential for clinical practical application.The designed Hi’CT system model will help the follow-up prototype design work.The principle analysis,the method analysis and system performance evaluation results for the Hi’CT system will also help to pro-vide important references in the subsequent prototype development and performance evaluation process.