Wu L, Gotman J
Montreal Neurological Institute and Department of Neurology and Neurosurgery, McGill University, Quebec, Canada.
Electroencephalogr Clin Neurophysiol. 1998 Apr;106(4):344-56. doi: 10.1016/s0013-4694(97)00156-9.
We present a method for the automatic comparison of epileptic seizures in EEG, allowing the grouping of seizures having similar overall patterns. Each channel of the EEG is first broken down into segments having relatively stationary characteristics. Features are then calculated for each segment and all segments of all channels of the seizures of one patient are grouped into clusters of similar morphology. This clustering allows labeling of every EEG segment. Methods derived from string matching procedures are then used to obtain an overall edit distance between two seizures, a distance that represents how the two seizures, taken in their entirety and including the channels not actually involved in the discharge, resemble each other. Examples from 5 patients, 3 with intracerebral electrodes and two with scalp electrodes, illustrate the ability of the method to group seizures of similar morphology.
我们提出了一种用于自动比较脑电图中癫痫发作的方法，该方法可对具有相似整体模式的发作进行分组。脑电图的每个通道首先被分解为具有相对稳定特征的段。然后为每个段计算特征，并将一名患者发作的所有通道的所有段分组为形态相似的簇。这种聚类允许对每个脑电图段进行标记。接着使用源自字符串匹配程序的方法来获得两次发作之间的整体编辑距离，该距离表示两次发作作为一个整体（包括未实际参与放电的通道）彼此的相似程度。来自5名患者的示例，其中3名患者使用脑内电极，2名患者使用头皮电极，说明了该方法对形态相似的发作进行分组的能力。
