Title :
Shrinking LORETA-FOCUSS: a recursive approach to estimating high spatial resolution electrical activity in the brain
Author :
Liu, He Sheng ; Yang, Fusheng ; Gao, Xiaorong ; Gao, Shangkai
Author_Institution :
Dept. of Biomed. Eng., Tsinghua Univ., Beijing, China
Abstract :
To image the neural current source from the measurements of EEG, this paper describes a new approach that integrates the ideas of Low Resolution Electromagnetic Tomography(LORETA) and Focal Underdetermined System Solution(FOCUSS). Basically a weighted minimum norm inverse method, this new algorithm-termed Shrinking LORETA-FOCUSS-recursively adjusts the weighting matrix and solution space until a sparse focal solution is achieved. The computer simulation compares its performance with LORETA and FOCUSS. The results suggest this method is able to reconstruct the 3D source distribution in brain volume with very small localization error and small energy error.
Keywords :
electroencephalography; iterative methods; least squares approximations; medical signal processing; source separation; 3-D source distribution; EEG inverse problem; Shrinking LORETA-FOCUSS algorithm; brain mapping; computer simulation; fast convergence; focal underdetermined system solution; high spatial resolution electrical activity; iterative step; low resolution electromagnetic tomography; neural current source; recursive approach; small energy error; small localization error; solution space; sparse focal solution; three-shell spherical head model; weighted MNLS algorithm; weighted minimum norm inverse method; weighting matrix; Computer errors; Computer simulation; Current measurement; Electroencephalography; Electromagnetic measurements; Image resolution; Inverse problems; Recursive estimation; Sparse matrices; Spatial resolution;
Conference_Titel :
Neural Engineering, 2003. Conference Proceedings. First International IEEE EMBS Conference on
Print_ISBN :
0-7803-7579-3
DOI :
10.1109/CNE.2003.1196884
