Improvement of resistivity images of wireless EIT systems with denoising approaches

In the last decade, Electrical Impedance Tomography (EIT) became an attractive issue leading to increased demand for various studies to be conducted in the field. In EIT, recorded voltages from different directions of test objects are captured and resistivity images are constructed. The recorded signals are affected by a number of noise sources coming from the surroundings of the laboratory, which impact the final resistivity images. In this study, we propose a wireless EIT system, which is capable of eliminating the effect of external noises from recorded signals. A mixture of normal saline and some inhomogeneities are made as a phantom. 10 electrodes have been used to inject the constant current to the phantom. For the image reconstruction, the electrode voltages are recorded and sent to the computer by the Bluetooth module as serial interface. In this paper three solutions for the noise reduction in recorded signals such as bandpass, adaptive filter and numerical integrator are recommended. Primitive results show that, these methods improved the quality of resistivity images and the inhomogeneity is significantly different from the other parts of the phantom. The accuracy of this apparatus is 97% for detecting of inhomogeneities from other segments.