Automatic identification of diffuse and localized activity for topographical distribution of EEG rhythm based on suitable reference selection with pre-judgments

An automatic reference selection method with prejudgments was developed in order to identify the two patterns of diffuse and localized EEG activity, as well as the topographical distribution. The diffuse activity is identified and the effect of adopted reference is analyzed by using two pre-judgments. The localized activity is analyzed where the focal and distributed electrodes are derived based on an iterative detection method. Finally, the suitable reference is determined which is adaptive to the actual characteristics of diffuse and localized activity. Additionally, the topographical distribution is obtained automatically during the reference selection process. The presented method has the performance to be an assistant and subjective tool for clinical application, especially for the quantitative interpretation on EEG rhythm.