Unveiling the Development of Intracranial Injury Using Dynamic Brain EIT: an Evaluation of Current Reconstruction Algorithms.
Haoting Li,Rongqing Chen,Canhua Xu,Benyuan Liu,Mengxing Tang,Lin Yang,Xiuzhen Dong,Feng Fu
DOI: https://doi.org/10.1088/1361-6579/aa8016
2017-01-01
Physiological Measurement
Abstract:OBJECTIVE:Dynamic brain electrical impedance tomography (EIT) is a promising technique for continuously monitoring the development of cerebral injury. While there are many reconstruction algorithms available for brain EIT, there is still a lack of study to compare their performance in the context of dynamic brain monitoring.APPROACH:To address this problem, we develop a framework for evaluating different current algorithms with their ability to correctly identify small intracranial conductivity changes. Firstly, a simulation 3D head phantom with realistic layered structure and impedance distribution is developed. Next several reconstructing algorithms, such as back projection (BP), damped least-square (DLS), Bayesian, split Bregman (SB) and GREIT are introduced. We investigate their temporal response, noise performance, location and shape error with respect to different noise levels on the simulation phantom. The results show that the SB algorithm demonstrates superior performance in reducing image error. To further improve the location accuracy, we optimize SB by incorporating the brain structure-based conductivity distribution priors, in which differences of the conductivities between different brain tissues and the inhomogeneous conductivity distribution of the skull are considered. We compare this novel algorithm (called SB-IBCD) with SB and DLS using anatomically correct head shaped phantoms with spatial varying skull conductivity. Main results and Significance: The results showed that SB-IBCD is the most effective in unveiling small intracranial conductivity changes, where it can reduce the image error by an average of 30.0% compared to DLS.