一种实用的动态电阻抗断层成象算法

A PRACTICAL RECONSTRUCTION ALGORITHM OF DYNAMIC ELECTRICAL IMPEDANCE TOMOGRAPHY

在电阻抗断层成象技术 (EIT)的研究中提出并推导了一种较为严格的图象重构算法 广义逆法 ,在此基础上针对EIT的病态性问题 ,对该算法进行了改进 ,并在一种新的计算机有限元模型上实现了该算法 ,重构出了反映阻抗变化量的图象 (即动态成象 )。改进的广义逆法突破了原广义逆法在剖分规模上的限制 ,能在较大的剖分规模上进行 ,更有利于图象的重构。将该算法用于计算机仿真研究 ,结果表明 :改进的广义逆法与现有算法 等位线法相比 ,成象定位更准确、重构图象的误差小 ,在信号信噪比较高时能较好地重构出阻抗分布复杂的模型。最后 ,对改进该算法的噪声性能进行了讨论。

A rigid electrical impedance tomography(EIT) reconstruction algorithm-general inversion algorithm was presented. To improve the ill-posed condition, we modified the algorithm and implemented it on a new FEM model, and get the image of impedance change(dynamic image). This modified general inversion algorithm(MGIA) could be implemented on a bigger dimension FEM model, so could do good to reconstruction. When implementing this algorithm on computer simulative condition, we found that the MGIA had a more accurate location ability and a smaller reconstruction error than the now-generally-used algorithm-equi-potential-back-projection algorithm. When having a relatively high SNR data, it could reconstruct images of complicated model. Finally, noise performance and improving method were discussed.