基于反向传播神经网络的阻抗断层图像重建新方法

A New Image Reconstruction Method for Electrical Impedance Tomography Based on BP Neural Network

电阻抗成像是通过对物体表面电压、电流的测量来重建物体内部阻抗分布或变化图像的一种新颖计算机断层成像技术。阻抗断层图像重建是一种病态的、非线性的逆问题。提出了一种全新的阻抗断层图像重建方法,它利用反向传播神经网络来表征物体内部阻抗变化位置与物体表面电压变化大小的非线性映射关系,从而可以根据对物体表面测量电压的变化先准确定位阻抗变化区域,再用线性近似方法重建阻抗变化图像。这种方法不仅具有一定的抗噪能力,而且成像精度和空间分辨率都大大好于逆投影方法。

Electrical impedance tomography (EIT) is a new computer tomography technology, which reconstructs an impedance (resistivity, conductivity) distribution or change by measuring voltage and current on the object' s periphery. Image reconstruction in EIT is an ill-posed, nonlinear inverse problem. A new image reconstruction method for EIT is proposed in which a BP neural network is used to express the nonlinear relation between the impedance change inside the object and the voltage change measured on the surface of the object. Thus, the location of the impedance change can be decided by the measured voltage change on the surface, and then the impedance change will be reconstructed with linear approximated method. This new image reconstruction method can alleviate some noise affection, and the precision and space resolution of the reconstructed image are greatly higher than that with the back projection method.