正电子发射断层扫描动态影像的智能型示踪剂动力学模型构建方法研究

项目名称： 正电子发射断层扫描动态影像的智能型示踪剂动力学模型构建方法研究

项目编号： No.61300150

项目类型： 青年科学基金项目

立项/批准年度： 2014

项目学科： 自动化技术、计算机技术

项目作者： 刘丽

作者单位： 江南大学

项目金额： 23万元

中文摘要： 正电子发射断层扫描是医疗和临床前研究中广泛使用的先进成像技术。本项目针对PET动态影像示踪剂动力学建模（dPET-TKM）获得具有生理学意义的新型医学影像定量指标这一科学问题，遵循药物代谢动力学理论及仿生免疫机理模型构建方法，采用理论研究与数值模拟、实验数据分析相结合的方法，系统开展智能型PET动态影像示踪剂动力学建模与算法研究。主要内容包括基于时间域分解和张量分析的高效示踪剂动力学模型数值求解，研究生物免疫树突状细胞抗原提呈功能及对免疫T细胞的激活调控机理，设计抽象模型复现关键激活调控规则，构建新型人工免疫网络模型。目标是建立一套基于人工免疫网络方法、高维形状和纹理模型以及张量分析相结合的智能型PET动态影像动力学特征提取和示踪剂动力学建模理论和方法，从而能够智能化分析处理复杂海量医学动态影像信息，为肿瘤等重大疾病早期诊断、药物研发，提供新型的定量影像探测方法，具有重要的社会和经济价值。

中文关键词： 正电子发射断层；动态成像；示踪剂动力学建模；人工免疫网络；反问题

英文摘要： Positron Emission Tomography has been widely used as a powerful tool for medical imaging, which plays an important role in many areas of medical treatments and preclinical researches. The purpose of this project is tracer kinetic modeling with dynamic PET (dPET-TKM) based on the pharmacokinetics principles and bio-inspired artificial immune system to generate new quantitative indicators of the medical images with physiological and biochemical functions. Theory studies, numerical simulations and experimental studies are combined in this project. Numerical solutions of tracer kinetic models are proposed to be obtained by the efficient time-domain decomposition and the tensor analysis. Inspired by the biological immune dendritic cell antigen-presenting function and the immune T-cell activation regulation mechanism,a new artificial immune network model is designed.More intelligent and more reliable methods for dPET-TKM will be implemented in this proposal, which support more complicated tracer kinetic models with the combination of artificial immune networks, high dimensional shape, texture models and higher tensor anslysis. Researches will not only improve intelligence and quantitative analysis of medical dynamic images, but also provides a new non-invasive method for both early diagnosis of major diseases and n

英文关键词： Positron Emission Tomography；Dynamic Imaging；Tracer Kinetic Modelling；Artificial Immune Network；Inverse Problem