摘要
ICU中,急性低血压的发生可能引起严重的后果,甚至威胁患者的生命安全,临床上主要依靠医生的经验进行预见性判断。为了实现急性低血压发生的自动检测和提前预报,本文运用医学信息学理论,对发生与未发生急性低血压两者间平均动脉压信号进行小波多尺度分解,并选取各层小波系数的统计特征参数中位数和最大值,用于支持向量机分类预测器的学习和训练,建立分类预测模型,预测准确率达90%。实验结果表明,该方法可以为ICU监护中急性低血压发生的提前预测和提前干预提供技术支撑,具有重要的临床应用价值。
The occurrence of acute hypotensive episode(AHE) in intensive care units(ICU) can result in grave consequences or even endanger the patients' lives.It is mainly depended on the clinical experience of doctors to treat.To detect automatically and forecast the occurrence of AHE,the theory of medical informatics has been applied in this paper.In our study,the mean arterial blood pressure(MAP) signals of those who experienced AHE and those who do not are both described on different scales by using wavelet transform.Through the extraction of the median and maximum from the wavelet coefficients for learning and training based on support vector machine(SVM),a predicting model with a predictive accuracy of 90% has been developed.The experiment demonstrates that the study of this approach is beneficial to early prediction of acute hypotension and intervention,which would significantly reduce the death risk of patients,with great value to clinical application.
出处
《透析与人工器官》
2011年第1期28-33,共6页
Chinese Journal of Dialysis and Artificial Organs
关键词
急性低血压
小波变换
特征参数
支持向量机
预测
acute hypotensive episodes
wavelet transform
parameters
support vector machine
prediction