求解光学CT图象重建问题的广义脉冲谱技术研究

间接法是目前实现医学诊断用近红外光三维成象(光CT)的最有潜力的手段之一.该方法基于以下假设,即给定分别对应于不同点激励源作用下成象组织体表面各点的传输光测量.在组织体内存在着与上述检测量相对应的、唯一的光学参数三维分布.由此,图象重建变成了特定光子传输模型的逆问题.本文提出采用广义脉冲谱技术(Generalized Pulse Spectrum Technique,GPST)进行光学CT图象重建逆问题的求解.给出了GPST在扩散方程光子传输模型逆问题求解中的具体实现,数值实验结果表明,通过选择合适的复频率点,采用GPST算法的图象重建结果优于其它现有算法,且响应时间合理.最后,本文还初步讨论了吸收和散射系数重建中复频率点选择的一些基本原则。

APPLICATION OF GENERALIZED PULSE SPECTRUM TECHNIQUE TO THE IMAGE RECONSTRUCTION IN OPTICAL COMPUTERIZED TOMOGRAPHY

Indirect methods have long been known as one of the most potential tools to implement 3D near infrared imaging for biomedical diagnosis.This approach is based on the assumption that,given a set of measurements of transmitted light between pairs of source and detection sites on the surface of an object,there exists a unique three dimensional distribution of internal absorptive and scattering parameters which would yield that set.Mathematically,the image reconstruction therefore becomes a task of solving an inverse problem to an appropriate model of photon transport.In this paper,it is presented a generalized pulse spectrum technique(GPST) for solving the inverse problem of the image reconstruction in optical tomography,and its implementation to a diffusion equation based inverse model.By properly selecting the working frequencies,the numerical expriments show that this algorithm has advantages over other existing methods of more accurate result and reasonable computational time.Finally,some principles in the choice of the working frequencies for the reconstruction are accessed.