论文部分内容阅读
癫痫是常见的神经系统疾病之一。癫痫发作的识别通常采用脑电测量记录中的癫痫发作起始点,以辅助医生进行诊断并对患者的发作状态报警。利用脑电信号的瞬态参数提出了一种自适应带宽特征,可用于提高癫痫发作检测精度。首先,利用经验模态分解(EMD)求得脑电信号的本征模态函数(IMF),并计算特定阶次IMF的解析信号;其次,利用该解析信号求解瞬时幅值与瞬时频率,对EEG信号的带宽特征添加权重,得到可用于癫痫检测的自适应带宽特征(Adaptive Bandwidth);最后,利用该特征完成癫痫发作检测。采用长达118 h 49 min的癫痫患者临床脑电数据进行实验,实验结果表明,自适应带宽特征的敏感性、特异性、准确性参数均比原特征取得明显提高。自适应带宽特征可提高癫痫发作检测精度并降低时间延迟,便于及时采取治疗措施,为临床检测提供了重要依据。
Epilepsy is one of the common neurological diseases. Epileptic seizure identification is usually based on the epileptic seizure starting point in EEG recording to assist the doctor in the diagnosis and alerting of the patient’s seizure status. Using the transient parameters of EEG signal, an adaptive bandwidth feature is proposed, which can be used to improve the detection accuracy of seizures. First, the EMD (EEG) is used to find the intrinsic mode function (IMF) of the EEG and calculate the analytic signal of the IMF of a particular order. Secondly, the instantaneous amplitude and the instantaneous frequency are solved by using the analytic signal The weight of EEG signal is added to obtain the adaptive bandwidth which can be used for epilepsy detection. Finally, the feature is used to complete the seizure detection. The clinical EEG data of epileptic patients up to 118 h 49 min were used for experiments. The experimental results show that the sensitivity, specificity and accuracy parameters of adaptive bandwidth feature are significantly improved compared with the original features. Adaptive bandwidth characteristics can improve the detection accuracy of seizures and reduce the time delay, facilitate the timely treatment and provide an important basis for clinical testing.