Evaluation by damped least-square algorithm on target identification ability of intracranial focal bidirectional electrical impedance perturbation / 医疗卫生装备

ABSTRACT

Objective To evaluate the effect of damped least-square algorithm on the identification of focal bidirectional electrical impedance perturbation in the intracranial area, and to analyze the influence of this kind of perturbation on the reconstruction quality of electrical impedance tomography. Methods Focal bidirectional electrical impedance perturbation was built based on the three-dimensional model and damped least-square algorithm was introduced into imaging. The position error and resolution were used to evaluate the imaging performance.Results When the focal electrical impedance perturbation existed,the target whose conductivity varied greatly or volume was large was more likely to be identified in the images while the target with smaller volume or variable conductivity presented in the reconstruction image was not obvious. It's pointed out that it may cause reconstruction images in some cases could not truly reflect the location and change information of the object of primary cerebral hemorrhage.At the same time,it affected the reconstruction precision,causing the position error and resolution with large fluctuation. Conclusion In reconstruction algorithm linearity correct matrix introduced for bidirectional electrical impedance perturbation contributes to enhancing the recognition of bidirectional perturbation, so that the characterization of the electrical impedance imaging can be augmented for clinical intracerebral hemorrhage.