Image Reconstruction for Electrical Impedance Tomography: Experimental Comparison of Radial Basis Neural Network and Gauss – Newton Method

Abstract Electrical impedance tomography (EIT) is an intensively researched noninvasive diagnostic method for medical use, that can help to improve the lung diagnostics, artificial lung ventilation and prevent lung injuries. Further improvements of reconstruction algorithms and measurement devices are essential to widen the use of EIT as a lung diagnostic method. To test potential of Radial Basis Neural Networks (RBNN) and Hopfield Neural Networks (HNN) for image reconstruction experiment is carried. Said neural networks are compared with Gauss – Newton (GN) algorithm. Results of the experiment show higher reconstruction accuracy with RBNN and HNN over GN algorithm.

[1]  T P Ryan,et al.  Temperature field estimation using electrical impedance profiling methods. I. Reconstruction algorithm and simulated results. , 1994, International journal of hyperthermia : the official journal of European Society for Hyperthermic Oncology, North American Hyperthermia Group.

[2]  Manuchehr Soleimani.  Electrical impedance tomography system: an open access circuit design , 2006, Biomedical engineering online.

[3]  M. Elia,et al.  Bioelectrical impedance analysis--part I: review of principles and methods. , 2004, Clinical nutrition.

[4]  Chein-I Chang,et al.  Robust radial basis function neural networks , 1999, IEEE Trans. Syst. Man Cybern. Part B.

[5]  Václav Snásel,et al.  Towards new directions of data mining by evolutionary fuzzy rules and symbolic regression , 2013, Comput. Math. Appl..

[6]  Sébastien Martin,et al.  On the influence of spread constant in radial basis networks for electrical impedance tomography , 2016, Physiological measurement.

[7]  Steffen Leonhardt,et al.  Assessment of regional lung recruitment and derecruitment during a PEEP trial based on electrical impedance tomography , 2008, Intensive Care Medicine.

[8]  Michal Prauzek,et al.  Impact of the Radial Basis Function Spread Factor onto Image Reconstruction in Electrical Impedance Tomography , 2015 .

[9]  Steffen Leonhardt,et al.  Electrical impedance tomography: the holy grail of ventilation and perfusion monitoring? , 2012, Intensive Care Medicine.

[10]  Tushar Kanti Bera,et al.  Improving Image Quality in Electrical Impedance Tomography (EIT) Using Projection Error Propagation-Based Regularization (PEPR) Technique: A Simulation Study , 2011 .

[11]  I. Frerichs,et al.  Gravity effects on regional lung ventilation determined by functional EIT during parabolic flights. , 2001, Journal of applied physiology.

[12]  Sebastien Martin,et al.  Nonlinear Electrical Impedance Tomography Reconstruction Using Artificial Neural Networks and Particle Swarm Optimization , 2016, IEEE Transactions on Magnetics.

[13]  Hamid Dehghani,et al.  A novel data calibration scheme for electrical impedance tomography. , 2003, Physiological measurement.

[14]  William R B Lionheart,et al.  GREIT: a unified approach to 2D linear EIT reconstruction of lung images , 2009, Physiological measurement.

[15]  F. Gleeson,et al.  The risks of radiation exposure related to diagnostic imaging and how to minimise them , 2011, BMJ : British Medical Journal.