摘要
背景:由于动物闪光视觉诱发电位本身易受干扰,信号微弱,在其特征提取方面的研究还鲜有报道。传统的平均叠加技术仅依靠增加叠加次数,对闪光视觉诱发电位提取效果不明显。目的:验证小波分析在慢性高眼压模型大鼠闪光视觉诱发电位特征提取中的可靠性。设计、时间及地点:自身对照实验,于2008-11/2009-03在首都医科大学动物科学部实验中心完成。材料:选用健康成年Sprague-Dawley大鼠20只,按随机数字表法分为造模6,8周组,每组10只。方法:大鼠单眼建立慢性高眼压模型;对侧眼不处理,作为自身对照眼。分别于造模后6,8周时进行双眼闪光视觉诱发电位测量。利用离散Meyer小波对原始信号进行7层小波分解,提取单一尺度小波系数,比较与原信号时域分析结果的一致性。对小波变换后的信号进行时域统计学分析。主要观察指标:小波变换后信号的P1,N2,P2潜时,N1P1,P1N2幅值。结果:闪光视觉诱发电位主要成分为N1P1N2P2N3五波。第四五层细节系数和(D4+D5)突出闪光视觉诱发电位局部快速变化的特点,波形规整,主要波形成分识别度高。对(D4+D5)进行统计学分析,造模后6周组造模眼与正常眼N1P1和P1N2眼间幅值差异均有显著性意义(P<0.05);造模组8周组各观察指标造模眼与正常眼间差异无显著性意义(P>0.05)。结论:小波分析可作为提取正常和模型大鼠闪光视觉诱发电位特征的可靠方法。
BACKGROUND： As animal flash visual evoked potential is sensitive to interference with weak signal. There have been rare reports about its feature extraction. Traditional averaging technique increases superposition times but not extract flash visual evoked potential. OBJECTIVE： To evaluate the reliability of the wavelet analysis on the feature extraction of rat flash visual evoked potential based on the chronic high intraocular pressure model. DESIGN, TIME AND SETTING： Self-controlled experiment was performed at the Laboratory Center of Department of Animal Science, Capital Medical University from November 2008 to March 2009. MATERIALS： A total of 20 healthy adutt Sprague-Dawley rats were selected and randomly divided into 2 groups （n=10）： 6-week model and 8-week model groups.
METHODS： The chronic high intraocular pressure model was on one eye of every rat; the other eye was not given any treatment as the self-control. Both eyes were given flash visual evoked potential test 6 and 8 weeks after modeling. The original data was decomposed into 7 levels by discrete Meyer wavelet. And the wavelet coefficient of each level was extracted to compare the consistency with time domain analysis results of the original signal. MAIN OUTCOME MEASURES： The latency of P1, N2, P2 and the amplitude of N1PI, PIN2 were measured.
RESULTS： The main components of the flash visual evoked potential were N1P1N2P2N3. Compared with the original signal, the sum of the forth and the fifth detail coefficients＇ reconstruction （D4＋D5） enhanced the recognition of the local change. There was statistical difference of （D4＋D5） in the amplitude of N1P1, P1N2 between the left and right eyes in the 6-week model group （P 〈 0.05）, but no significant differences were found between both eyes in the 8-week model group （P 〉 0.05）.
CONCLUSION： Wavelet analysis can be used as one of the reliable methods to extract the flash visual evoked potential of normal and model rats.
出处
《中国组织工程研究与临床康复》
CAS
CSCD
北大核心
2009年第22期4287-4290,共4页
Journal of Clinical Rehabilitative Tissue Engineering Research
基金
国家自然科学基金资助项目(30670514,10802053)
北京市教委重点项目(KZ200710011010)
北京市属市管高等学校“学术创新团队计划”项目~~
