论文部分内容阅读
目的:实现室颤信号与非室颤信号的分类,进而实现室颤信号的检测。方法:本文引入了一种基于支持向量机(Support Vec-tor Machine,SVM)和改进的越限区间算法(TCI)的新算法,其中支持向量机在处理分类和模式识别等问题中具有很大的优势。该算法采用4s的滑动窗技术,并利用改进后的越限区间算法(Threshold Crossing Interval,TCI)方法提取心电信号的特征。新算法的实现如下:在每一滑动窗内采用改进的后的绝对值阈值,计算中间2s内的平均越限间隔值。并以此TCI值作为特征参数,输入一个预先设计好的二分类支持向量机中,从而实现分类。结果:成功实现了室颤信号的检测,通过计算该方法的灵敏度、精确度、预测性和准确度且与其他方法相比较,可知此新算法总体可靠性优于其他方法。结论:该算法能够实现室颤信号的实时监测,且简单易行,易于实现,较适合实时的心电监测以及除颤仪器。
Objective: To realize the classification of ventricular fibrillation signals and non-ventricular fibrillation signals, and then realize the detection of ventricular fibrillation signals. Methods: This paper introduces a new algorithm based on support vector machine (SVM) and improved TCI algorithm, in which SVM has a great deal of problems in dealing with classification and pattern recognition The advantages. The algorithm adopts 4s sliding window technique and extracts the characteristics of ECG signal by using the improved Threshold Crossing Interval (TCI) method. The new algorithm is implemented as follows: In each sliding window, the improved absolute threshold is used to calculate the average threshold interval in the middle 2s. And take this TCI value as the characteristic parameter, enter a pre-designed two-class SVM, in order to achieve classification. Results: The detection of ventricular fibrillation signal was successfully achieved. Compared with other methods, the overall reliability of the new algorithm was better than other methods by calculating the sensitivity, accuracy, predictability and accuracy of the proposed method. Conclusion: The algorithm can realize the real-time monitoring of ventricular fibrillation signal and is simple and easy to implement. It is suitable for real-time ECG monitoring and defibrillation apparatus.