Cross-patient epilepsy electroencephalogram signal classification method based on data enhancement and adversarial optimization

The invention relates to a cross-patient epilepsy electroencephalogram signal classification method based on data enhancement and adversarial optimization, belongs to the technical field of epilepsy detection, and solves the problems of low accuracy and efficiency of an epilepsy cross-patient electroencephalogram signal classification method in the prior art. The method comprises the following steps: acquiring electroencephalogram signal fragments, performing data enhancement on the electroencephalogram signal fragments in an epileptic seizure period, and constructing a training sample set based on the data-enhanced electroencephalogram signal fragments; constructing a cross-patient epilepsy electroencephalogram signal classification model, and training the cross-patient epilepsy electroencephalogram signal classification model based on the training sample set by adopting an adversarial optimization strategy to obtain a trained cross-patient epilepsy electroencephalogram signal classification model; and input