Liyun Rao

A novel combination method of electrical impedance tomography inverse problem for brain imaging

A novel method combined differential evolution algorithm and modified Newton-Raphson method is proposed in this paper to solve the electrical impedance tomography (EIT) inverse problem. This method is applied to the two-dimensional impedance reconstruction of brain section based on a four-layer concentric circle model and real head geometric model. Our simulations demonstrate that the novel combination method is robust and time saving in obtaining high-quality reconstruction in EIT problems for brain imaging studied in this paper.

An efficient improvement of modified Newton-Raphson algorithm for electrical impedance tomography

An efficient improvement, based on the idea of homotopy, is proposed to improve and ensure the convergence of the modified Newton-Raphson (MNR) algorithm through the continuous mapping of solution space, and the behavior of solution is restrained towards the global convergence after several initial solution mappings. Comparisons of the new algorithm with MNR are presented, and the advantage of new algorithm is demonstrated for the problem of electrical impedance tomography.

Image reconstruction of EIT using differential evolution algorithm

Differential evolution (DE) algorithm is used in this paper to solve the inverse problem of EIT, where the cost function is determined by solving the forward problem using finite element method (FEM). This method is applied to the 2D impedance reconstruction of brain section based on real head model. Our simulations demonstrate that DE algorithm is robust in obtaining high quality reconstruction for EIT problems studied in this paper.

The method of maximum mutual information for biomedical electromagnetic inverse problems

This paper proposes a new method for biomedical electromagnetic inverse problems based on the technique of maximum mutual information. The new method is different from the conventional methods for it will not depend on the widely used regularization technique. The new method provides a general paradigm for developing algorithms dealing with various biomedical inverse problems that can be described using lead field, including localization and imaging of neural source activities in brain and heart from EEG/MEG and ECG/MCG; it is also possible to apply it to other, electromagnetic inverse problems like electrical impedance tomography. This paper mainly provides the theoretical development, with MEG inverse problems as case studies.

Mean field annealing (MFA) and optimal design of electromagnetic devices

This paper presents an optimal design method by means of mean field annealing (MFA). The mean field theory (MFT) is introduced, and a certain MFA application extension based on Peierls inequality is explained in detail, the critical temperature that is important for classes of problem while applying MFA is discussed. With MFA, the optimal size design of electromagnetic device is carried out. The results are compared with two existing algorithms, the modified simulated annealing (MSA) and one-variable stochastic simulated annealing based MFA (S-MFA), in both operating cost and final quantities. To modify the final results to reach the global optimum, Hooke-Jeeves pattern search method is adopted. Favorable results show the effectiveness of the proposed scheme.

Computations of electroencephalography and magnetoencephalography for real head model

Methods for constructing a real head model from MRI (magnetic resonance imaging) data and its boundary element mesh are proposed. Based on this real head model, forward computations of electroencephalography and magnetoencephalography are investigated and numerical simulation results are presented. Deflation is adopted in order to solve the singularity of discrete system equations in the boundary element method. An isolated problem approach is applied to overcome the smear effect of skull conductivity. An improved auto solid angle algorithm is developed to speed up the computation of kernel matrix.

EEG source localization using differential evolution method

Differential evolution (DE) method is used in This work to solve the EEG source localization problem based on equal current dipole model. The single dipole sources with four-shell concentric sphere model are reconstructed. Our simulations demonstrate that DE algorithm is robust in obtaining high quality reconstruction for EEG problems with single current dipole sources.

The Application of Magnetic Resonance Perfusion Imaging in the Estimation of Brain Function Using SVD Method

MR perfusion imaging can be used to estimate parameters indicating the metabolic process, including cerebral blood flow (CBF). Specifically singular value decomposition (SVD) method is applied in this paper to obtain CBF with a predefined conventional arterial input function (AIF), and the effects of threshold and tracer delay are investigated. While the simulation results show that SVD method can estimate CBF with good accuracy by using different thresholds for different flow values, the method is found to be sensitive to tracer delay. A delay correction scheme is advocated, where CBF is determined by SVD after time-shifting of the tracer concentration curve. With the correction the simulation results are considerably improved in the flow estimation.

Three EIT approaches for static imaging of head

Three EIT approaches for static imaging of head are investigated in this paper. The modified Newton-Raphson (MNR) method and the differential evolution (DE) algorithm are applied to the impedance reconstruction of 2D section of head based on real head model. Comparisons are carried out on the results obtained using simulated data, and a DE-MNR combination method is proposed, which demonstrated high impedance reconstruction quality with fast convergence in the 2D EIT simulation for static imaging of head.

MNR Method with Self-Determined Regularization Parameters for Solving Inverse Resistivity Problem

The modified Newton-Raphson (MNR) method is used to solve the inverse resistivity problem in this paper. Using Tikhonov regularization method, comparisons among the L-curve method, the zero-crossing (ZC) method and the generalized cross validation (GCV) method are carried out for determining the regularization parameters of MNR method. By these criterions the appropriate regularization parameters are self-determined and adjusted with the reconstruction iterations. Our simulation experiments on 2D circle model showed that the GCV method can provide the best reconstruction quality with the fastest speed in inverse resistivity problem using MNR method.