Implementation of 2D Electrical Impedance Tomography Image Reconstruction
Hansen Leonard Andreas, Farkhad Ihsan Hariadi, Hasballah Zakaria
School of Electrical Engineering and Informatics, Institut Teknologi Bandung, Jln. Ganesha 10, Bandung 40132, Indonesia . e-mail : hariadi[at]stei.itb.ac.id
Abstract
Abstract - Electrical Impendance Tomography is a method of reconstructing image pieces of an object by utilizing the objects conductivity feature. Image reconstruction is the last module of data processing on EIT after data stimulation and acquisition. At this stage used software MATLAB 2013a equipped with EIDORS v3.8 as the main tool. The main algorithm used is Forward Solver to convert the conductivity distribution value into boundary voltage and Inverse Solver which converts the boundary voltage value into conductivity distribution. In this module is used Finite Element Model as a mapping method on the observed spatial domain. Overall, the algorithm used refers to the Gauss-Newton Mathematical Minimization Algorithm algorithm coupled with NewtonRaphson Iteration to update the conductivity. The process will end when the difference between the FS yield voltage and the measuring voltage is below a tolerance value. In addition to image reconstruction,this module also comes with PC UART communication with FPGA using MATLAB and simulation on EIDORS. The additional feature of this module is to perform image reconstruction with 16 and 32 electrodes with manual inter-object separation with hyperparameters manually
Keywords: EIT, EIDORS, Forward Solver, Inverse Solver, Image Reconstruction
Topic: Biomedical, Robotic and ICT engineering