Characterizing neuronal firing patterns in the human brain.

A procedure to characterize firing patterns of neuron spikes from the human brain, in both temporal domain and the frequency domain, is presented. The combination of multitaper spectral estimation and the polynomial curve-fitting method is employed to transform the firing patterns to the frequency domain. To generate temporal shapes, eight local maxima are smoothly connected by cubic spline interpolation. We then used a rotated principal component analysis, which removes the orthogonality constraints of traditional PCA, to extract common firing patterns as templates from around 4100 neuron spike signals. Dynamic time warping was used to assign each neuron firings into the closest template without shift error. This technique can be utilized for finding firing similarities in neuroscience applications and in development of a query system.