The inverse problem of Electrical Impedance Tomography (EIT) is non-linear and severely ill-posed, resulting in relatively low image quality, which specifically, involves the aspects of structure preservation and conductivity contrast differentiation. In this paper, we report a kernel method based multi-modal EIT image reconstruction approach to tackle this challenge. The kernel method performs image-level segmentation-free information fusion and incorporates the structural information of an auxiliary high-resolution image into the EIT inversion process through the kernel matrix, which leads to an unconstrained least square problem. We describe this approach in a general way so that the high-resolution images from a variety of different imaging modalities can be adopted as the auxiliary image, if they contain sufficient structural information. In comparison with some state-of-the-art algorithms, the proposed kernel method generates superior EIT images on challenging simulation and experimental phantoms. Moreover, it presents the advantage of suppressing the interference of the existence of imaging-irrelevant objects in the auxiliary image to some extent. Simulation and experiment results also suggest the kernel method has great potential to be applied to more complicated tissue engineering applications in the future.
|Journal||IEEE Transactions on Instrumentation and Measurement|
|Publication status||Accepted/In press - 23 Nov 2021|