数据丢失情况下脑电图记录中样本熵与去趋势波动分析性能的比较研究
Comparative study between Sample Entropy and Detrended Fluctuation Analysis performance on EEG records under data loss.
作者信息
Cirugeda-Roldán E M, Molina-Picó A, Cuesta-Frau D, Oltra-Crespo S, Miró-Martínez P
机构信息
Computer Science Department (DISCA) at Polytechnic University of Valencia, Alcoy Campus (EPSA-UPV), 03801 Alcoy, Alicante, Spain.
出版信息
Annu Int Conf IEEE Eng Med Biol Soc. 2012;2012:4233-6. doi: 10.1109/EMBC.2012.6346901.
This study compares two signal entropy measures, Sample Entropy (SampEn) and Detrended Fluctuation Analysis (DFA) over real EEG signals after a randomized sample removal. Both measures have demonstrated their ability to discern between, among others: control and pathologic EEG signals, seizure free or not, control and opened eyes EEG, and side of brain signals. Results show that SampEn behaves better when analyzing control signals, while DFA provides better segmentation results between epileptic signals, in the context of sample loss, particularly when discerning between seizure and seizure free signal intervals.
本研究比较了两种信号熵测量方法，即样本熵（SampEn）和去趋势波动分析（DFA），用于分析随机去除样本后的真实脑电图（EEG）信号。这两种测量方法都已证明它们能够区分多种情况，包括：对照和病理性EEG信号、有无癫痫发作、对照和睁眼EEG以及脑信号的侧别。结果表明，在分析对照信号时，样本熵表现更好，而去趋势波动分析在样本丢失的情况下，特别是在区分癫痫发作和无癫痫发作信号间隔时，能提供更好的癫痫信号分割结果。
Zotero 插件