动态感应电流电阻抗断层成像的算法仿真  

Algorithm of dynamic induced current electrical impedance tomography based on simulation phantom

在线阅读下载全文

作  者:马建英[1] 董秀珍[1] 秦明新[1] 刘锐岗[1] 尤富生[1] 向海燕[1] 

机构地区:[1]第四军医大学生物医学工程系医学电子工程教研室,陕西西安710033

出  处:《第四军医大学学报》2002年第8期752-755,共4页Journal of the Fourth Military Medical University

基  金:中国 -以色列国际合作项目 (99M-0 0 2 14 2 6)

摘  要:目的 在仿真模型的基础上 ,研究一种用感应电流激励的动态电阻抗断层成像算法—牛顿迭代法的特性及其对独立测量数的依赖性 .方法 利用有限元法对成像区域进行离散 ,然后再用牛顿迭代法针对不同的线圈数进行求解、成像 ,以资比较 .结果 对同一目标 ,分别就不同的线圈数得到了成像结果 ,表明牛顿迭代法对电导率扰动的定位是基本准确的 ,成像误差随线圈数的增加而减小 .结论 用牛顿迭代法解动态感应电流电阻抗断层成像的逆问题是可行的 ,但在独立测量数小于剖分单元数的情况下 ,迭代过程不会准确的收敛于实际的电导率分布 ,而是一种近似 ;在独立测量数大于剖分单元数的情况下 ,迭代过程可以收敛于实际的电导率分布 。AIM To study the property of the algorithm of dynamic induced current electrical impedance tomography Newton iteration method and its dependency on independent measurement. METHODS FEM method was used to discretize the imaging area, and then Newton iteration method was used to solve the problem and reconstruct images with different numbers of coils for comparision. RESULTS With different coils, different images were obtained. The results indicated that Newton iteration method could be used to locate the conductivity perturbation accurately, and that when the number of independent measurement increased, the imaging error would decrease. CONCLUSION Newton iteration method is a feasible method for the inverse problem of dynamic induced current electrical impedance tomography. But, when the number of independent measurement is less than the number of FEM element, the iterative process will not converge accurately at the practical conductivity and the result is just an approximate. When the number of independent measurements is larger than the number of FEM element, the iterative process can converge accurately at the practical conductivity, by which high quality images will be obtained.

关 键 词:动态感应电流 电阻抗 断层成像 算法 仿真 计算机模拟 牛顿迭代法 

分 类 号:R318.03[医药卫生—生物医学工程]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象