基于邻域信息和快速FCM的肺部电阻抗成像伪迹优化算法

针对电阻抗成像技术可视化过程中因“欠定”问题和“软场”效应所导致的重建图像伪迹问题，该文提出一种基于邻域信息和快速模糊C均值聚类(快速FCM)的无监督图像质量评价指标。基于该评价指标和Tikhonov正则化算法，提出了一种重建图像伪迹优化算法TR-NC。仿真结果表明，该算法能够有效地修正重建图像中的伪迹，修正后的重建图像的相关系数平均提高了18.45%，相对误差平均降低了22.2%；仿真体验实验结果表明，当目标电导率变化率大于30%时，该算法能够准确地检测到目标。由此可见，相比于传统的Tikhonov正则化算法，提出的修正算法在重建图像目标的数量和位置精确度方面都得到了显著提高，为电学层析技术在医学和工业等领域的应用实践提供了新的成像理论依据和技术参考。

Artifact Optimization Algorithm for Pulmonary Electrical Impedance Tomography Based on Neighborhood Information and Fast FCM

To solve the problem of reconstruction image artifacts caused by the problem of "underdetermined" and the “soft field“ effect in the visualization process of electrical impedance tomography, an unsupervised image quality evaluation index based on neighborhood information and fast Fuzzy C-Means clustering (fast FCM) is proposed. Based on this evaluation index and Tikhonov regularization algorithm, a reconstruction image artifact optimization algorithm TR-NC is proposed. Simulation results show that the proposed algorithm can effectively correct artifacts in the reconstructed image, and the correlation coefficient of the modified reconstructed image has increased by 18.45% on average, and the relative error has reduced by 22.2% on average. Simulation experimental results show that the proposed algorithm can accurately detect the target when the change rate of target conductivity is more than 30%. It is shown that compared with the traditional Tikhonov regularization algorithm, the proposed modified algorithm TR-NC has been significantly improved in the number and position accuracy of reconstruction image targets, which provides a new imaging theoretical basis and technical reference for the application of electrical tomography technology to medical and industrial fields.