Non linear state estimation using the unscented Kalman filter in electrical impedance tomography

• Grant number: 08/07150-0
• Support type: Scholarships in Brazil - Doctorate (Direct)
• Effective date (Start): November 01, 2008
• Effective date (End): October 31, 2012
• Field of knowledge: Engineering - Biomedical Engineering - Medical Engineering
• Principal researcher: Raul Gonzalez Lima
• Grantee: Fernando Silva de Moura
• Home Institution: Escola Politécnica (EP). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Associated research grant: 01/05303-4 - New strategies in artificial ventilation: diagnosis and prevention of barotrauma/biotrauma through electrical impedance tomography (EIT), AP.TEM

Abstract

The Electrical Impedance Tomography (EIT) is a method, which tries to supply the observation necessity of an impedance distribution in a closed domain. One of the EIT applications is the real time monitoring of the lungs conditions in patients in intensive care units because one can find relation between its conditions with the electrical properties of tissues and atmospheric gases. Among various algorithms to solve this problem, are the Kalman filters, which address the estimating problem with a probabilistic point of view through a method of minimum variance of estimation error. To use these filters, one temporal evolution model of the system must be used. Due to the complexity of the lungs dynamics, until today the only model used is the random walk model. This research aims to propose dynamic models for the main tissues involved in the respiratory cycle and use them in the Unscented Kalman Filter (UKF), a non-linear Kalman filter extension not yet explored in the electric impedance tomography. The models will be adjusted in real time by increasing the system's state, including the model parameters in the estimated variables. The decision on which model to adopt in the different regions of the chest will be made by a segmentation algorithm. The tests will be carried out through numerical simulations of a time varying resistivity distribution in a domain and through data already collected from animals. (AU)

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

MARTINS, THIAGO DE CASTRO; GUERRA TSUZUKI, MARCOS DE SALES; BUENO DE CAMARGO, ERICK DARIO LEON; LIMA, RAUL GONZALEZ; DE MOURA, FERNANDO SILVA; PASSOS AMATO, MARCELO BRITO. Interval Simulated Annealing applied to Electrical Impedance Tomography image reconstruction with fast objective function evaluation. COMPUTERS & MATHEMATICS WITH APPLICATIONS, v. 72, n. 5, p. 1230-1243, SEP 2016. Web of Science Citations: 3.