Automatic Discrimination of Abnormal SubjectsUsing the Visual Evoked Potential Spectral Components

[摘要] Study of visual evoked potential (VEP) is one of the utilizedmethods in clinical diagnosis of ophthalmology and neurologicaldisorders. The automatic detection of VEP spectral components isan important tool in the diagnosis of mental activity. This paperpresents a novel computational approach using feedforward neuralnetwork to identify abnormal subjects from changes in spectralcomponents. The output vector from the feedforward neural networkis based on the VEP spectral components. The software wasdeveloped to identify mental state from the VEP spectralcomponents using Matlab software package. Using this approach, itis possible to perform real-time abnormality identificationaccurately on personal computers.

[学科分类] 基础医学