Optimized Method for Electrical Impedance Tomography to Image Large Area Conductive Perturbation

Abstract

The objective of the study was to develop an optimized method for dynamic electrical impedance tomography (EIT) to image large area conductive perturbation (LACP), a new type of imaging target that we found during the monitoring and evaluation of the mannitol dehydration treatment of patients with brain edema based on brain EIT. Previously, we reconstructed LACP images with the commonly used NOSER algorithm with polar driven pattern. However, conductivity changes near the center of the LACP were blurred or remained undetected, and blocky artifacts appeared in the reconstructed images, making the interpretation of the results difficult. To solve this problem, we, for the first time, propose an optimized algorithm for imaging LACP. This algorithm comprises a modified sensitivity matrix to compensate for the blurred conductivity changes near the center. It also uses the Markov random field constraint to reduce blocky artifacts. To verify the performance of the proposed method, we conducted experiments based on head models and human subjects. Specific metrics, including shape error (SE) and image fluctuation (IF) artifacts, were also proposed to evaluate the image quality of the LACP. The experimental results demonstrated that, compared with the NOSER algorithm, the proposed method could respectively reduce SE and IF by 51.7% and 47.5%. Therefore, it can optimize the imaging of LACP and provide references for other applications at present or in the future which involve imaging LACP.

Keywords

• Electrical impedance tomography
• large area conductive perturbation
• sensitivity matrix
• brain structure
• reconstruction algorithm