摘要
目的研究表征血压波形间相似程度的量化指标,并应用到脉动流模拟系统中血压波形的分析。方法在对已有相似度算法进行研究的基础上,结合血压波形自身的特点,提出加权平均算法。该算法将相关系数法和特征参数法相结合,分别从整体和局部特征对波形进行相似性度量,兼具全局匹配和局部比较的优点。结果与夹角余弦法、平均绝对差法和数值型相似系数法相比,加权平均算法更适用于血压波形相似度的分析。结论加权平均算法可以表征不同血压波形之间的差异,比较不同脉动流装置之间的模拟性能,进一步完善后还可以用于其他生理波形的分析。
Objective To study a quantitative indicator for measuring the similarity between blood pressure waveforms and its application in the analysis on the simulated blood pressure waveform for pulsatile flow simulation system.Methods Based on the past similarity measurement algorithm and the known feature of blood pressure waveforms,the weighted average algorithm was presented in this paper,which possessed advantages of both global matching and partial matching by integrating the correlation coefficient with the characteristic parameter algorithm,and calculating similarity degrees through overall and partial waveform.Results The weighted average algorithm was proved to be more suitable for the analysis on the blood pressure waveform similarity compared with the angle cosine method,average absolute deviation method,and numeric similarity coefficient method.Conclusions The weighted average algorithm showed excellent ability in calculating the similarity degree between different waveforms,or in comparing the performance with different pulsatile flow simulation systems,and it could be applied to other physiological waveforms with further improvement.
出处
《医用生物力学》
EI
CAS
CSCD
北大核心
2012年第5期515-520,共6页
Journal of Medical Biomechanics
基金
国家自然科学基金资助项目(10925208)
关键词
相似度
相关系数
特征参数
血压波形
脉动流
生物反应器
波形分析
Similarity
Correlation coefficient
Characteristic parameter
Blood pressure waveform
Pulsatile flow
Bioreactor
Waveform analysis