论文部分内容阅读
Objective:The study presents a multi-channel patient-specific seizure detection method based on theEmpirical mode decomposition (EMD) and support vector machine (SVM) classifier.Method: The Empirical mode decomposition (EMD) is used to extract features fromintracranialEEG.A machine-learning algorithm (SVM) is used as a classifier to discriminate between seizure and non-seizure intracranialEEG epochs.A post-process ing algorithm is proposed to reject artifacts and increase the robustness of the method.The proposed method was evaluated using482 h of intracranialEEG recordings from 17 patients with total of 51 seizuresin the Freiburg EEG database.Results: The proposed method showed a better performance than most of the existing seizure detection systems.The method was able to achieve an average sensitivity of 92%, a false detection rate of 0.17/h, and a time delay of 12s.Moreover, the false de tection rate could be further reduced by extension of the time delay.Conclusions: Given high sensitivity and low false detection rate, the proposed patient specific seizure detection method can greatly assist clinical staff to mark seizuresin long-term EEG automatically or detect seizure onsetonline with high performance.Early and accurate seizure detection using this method may provide a practical tool for planning epilepsy intervention.