基于射野影像的剂量重建方法之可行性初探

利用治疗时得到的影像信息重建病人体内所接受的真实剂量是剂量引导下放射治疗（DGRT）技术临床应用的重要内容之一。治疗剂最重建方法之一是：通过匹对模拟的射野影像和治疗时获得的射野影像来反推射野注量分布，然后计算出治疗时病人体内实际接受的剂量：由此重建出的射野注量是否精确受诸多因素的影响．因素之一是是否能真实地模拟射野影像。本文用傅立叶卷积积分法模拟射野影像，采用不同积分半径得剑不同程度散射近似下的模拟射野影像。通过用不同程度散射近似下的模拟射野影像来反推射野注量分布．考察了散射因素对反推射野注量分布的影响。发现：即便射野影像在没有噪音和分辨率足够高的理想情况下，要精确地反推射野注量，必须在模拟射野影像中充分考虑散射因素。

Feasibility Study of a Dosimetric Verification Method Based on EPI

To reconstruct the delivered dose in a patient by utilizing the image information acquired during treatment is an important application of IGRT techniques. One of the approaches for the reconstruction of the delivered dose is to : Firstly, deduce the incident fluence distribution through matching the simulated portal image and an acquired portal image during treatment. Then calculate the dose distribution in the patient. Many factors play roles in the reconstruction approach and thus affect the accuracy of final results. One of these factors is whether the portal image is simulated appropriately. This article use the convolution/superposition algorithm to simulate the portal image. A set of simulated portal images of different scattering approximation are obtained by using different integral radius in the convolution/superposition algorithm. By using these simulated portal images to reconstruct the incident fluence distribution of the radiation field,the influence of the scattering factor is investigated. We find that an accurate reconstruction of the fluence distribution need to sufficiently consider the scattering in the portal image simulation even though in a ideal case (resolution high enough, no noise and so on).