High resolution imaging of bioelectric sources

The present study investigates various techniques for high resolution imaging of brain and heart electrical sources from non-invasive electroencephalography (EEG) and electrocardiography (ECG) measurement.; The surface Laplacian technique, as a spatial enhancement method, was applied to both neural and cardiac imaging. A novel spline Laplacian algorithm was developed to take into consideration of the realistic geometry of the body surface, and its excellent performance was demonstrated by both computer simulations and human experiments.; High-resolution neuroimaging was also achieved by solving the EEG inverse problem. Different regularization techniques were explored in order to improve the numerical stability of the inverse problem, and a novel minimal product method was proposed for easy and rational determination of the regularization parameter.; The cortical imaging technique was investigated for high resolution mapping of brain activity over the brain surface. A forward theory on the equivalent cortical current dipole layer was developed, which provides theoretical justification on the cortical current imaging approach. The cortical potential imaging was successfully applied to cognitive neuroimaging studies.; The three-dimensional neuroimaging was also explored by solving equivalent current source distribution within three-dimension of the brain volume. Compared to the conventional equivalent dipole source model, the new method reduces the dimension of the solution space, thus improving the computational efficiency, reducing the under-determination of the inverse problem, and enhancing the stability of the inverse solution.