J.C. Mosher and R.M. Leahy, "Source localization using recursively applied and projected (RAP) MUSIC, " IEEE Trans. Signal Process., vol.47, no.2, pp.332-- 340, Feb. 1999.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....: VXW BYO . After placing dipoles at each peak, we can easily determine the direction of each dipole and its magnitude by solving linear optimization problems. 4. UFO MUSIC METHOD Although the expansion of the MUSIC method, e.g. the recursively applied and projected MUSIC (RAP MUSIC) method [4], represents significant advances in source localization, it still fails to recognize the unreliability of multiple dipole source solutions as compared to single dipole source solutions. These solutions are very sensitive to noise, which limits the usefulness of these solutions from a clinical ....

....Although ICA and the analysis of the MEG inverse problem have been separately discussed [9] 10] the proposed method incorporates ICA into the MEG inverse solution. However, when sources are correlated or synchronous, the proposed UFO MUSIC may have to be expanded to a RAP MUSIC like method [4]. Further studies will clarify how to approach such a case. 7. ACKNOWLEDGEMENT We thank E.Rosenberg for valuable comments on this manuscript. 8. ....

J.C. Mosher and R.M. Leahy, "Source localization using recursively applied and projected (RAP) MUSIC, " IEEE Trans. Signal Process., vol.47, no.2, pp.332-- 340, Feb. 1999.

....and the potential be reconstructed at the discretisation points. Since these are usually more numerous than the electrodes, the problem is highly underdetermined. Due to its illposed ness regularization techniques have to be employed. We refer the reader e.g. to [HH93, K oh98, DSB01, PM99, ML99] All three approaches involve in one way or the other the setup up a so called lead eld matrix. Lead Field Matrix Assume that a dipole is given with position r and orientation d and let E be a set of N electrodes. The vector M( r; d ) of potentials generated by the dipole at the ....

J. Mosher and R. Leahy. Source localization using recursively applied and projected (RAP) MUSIC. IEEE Transactions on Signal Processing, 47(2):332-340, 1999.

....brain regions is only the rst step in demonstrating the usefulness of BSS algorithms. The next step is to systematically study the e ect of BSS on source localization when BSS methods are combined with more sophisticated source localization algorithms (Ribary et al. 1991; Aine et al. 1998; Mosher and Leahy, 1999)) Through the application of a BSS algorithm to MEG data, we have previously shown that (1) BSS is capable of separating various artifacts from neuronal sources (Tang et al. 1999) 2) BSS is capable of separating neuronal sources at di erent processing stages along the visual pathways; and (3) ....

Mosher, J. and Leahy, R. (1999). Source localization using recursively applied projected (RAP) MUSIC. IEEE Transactions on Signal Processing, 47(2):332-340.

....that are described here, these relationships must be expressed in matrix form. In [15] we reviewed matrix forms of the lead field [4] 29] for EEG and MEG measurements, for both # In [11] 17] we referred to R MUSIC as RAP MUSIC; we have since adopted the terminology used here and in [18] to differentiate the two methods. Mosher, Leahy: R MUSIC Version 2 Page 6 of 40 Los Alamos Technical Report # LA UR 97 3182 July 31, 1997 spherical and general BEM head models. In each case, the measurements can be expressed as an explicit function of primary current activity; the passive ....

Mosher JC, Leahy RM, "Source localization using recursively applied and projected (RAP) MUSIC," Los Alamos National Laboratory Technical Report LA-UR-97-1881.

....vectors associated with each of the identified event related temporal components to construct a signal subspace. This follows directly by analogy to PCA, where the spatial components are used to construct a signal subspace from which focal current source locations are found using a MUSIC algorithm [3]. The attraction of using ICA methods for signal subspace identification is that they can be applied directly to raw (unaveraged) E MEG data and do not require that the noise components in the data be spatially uncorrelated. Conversely, SVD based identification of the signal subspace assumes that ....

....signal subspace identification is that they can be applied directly to raw (unaveraged) E MEG data and do not require that the noise components in the data be spatially uncorrelated. Conversely, SVD based identification of the signal subspace assumes that any noise in the data is spatially white [3]. Here we investigate the utility of ICA for estimating the signal subspace and compare its performance with that of the conventional SVD based methods using both raw and averaged data. In using ICA for signal subspace selection, we must identify which components are event related. Typically, ....

[Article contains additional citation context not shown here]

J.C. Mosher and R.M.Leahy, "Source localization using recursively applied and projected (RAP) MUSIC", IEEE Trans Sig. Proc. 47(2), 1999.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC