电子胃镜运行状态异常智能告警方法研究

Research on intelligent alarm method of abnormal operation state of electronic gastroscope

针对电子胃镜运行状态异常告警中普遍存在检测率低的问题,提出一种新的电子胃镜运行状态异常智能告警方法。通过双重马尔可夫嵌入状态过程对电子胃镜运行状态的正常或异常行为进行描述,建立正常状态HMM模型,利用Baum-Weltch算法完成HMM学习,递归估计HMM模型参数,令重新得到的HMM参数针对给出观测序列的后验概率逐渐升高直到收敛,完成对训练样本序列的有效解释,针对存在的评估问题、解码问题、学习问题进行解决。在对电子胃镜运行状态进行异常智能告警的过程中,利用滑动窗口形成的短序列对电子胃镜当前行为进行描述。针对描述正常电子胃镜运行状态的隐马尔可夫模型参数,通过求解输出概率对该序列相对于正常模型行为的偏差进行度量,偏差越大,认为该行为处于异常状态的可能性越大,实现异常智能告警。经验证,所提方法告警准确,整体性能优。

In order to solve the problem of low detection rate in the abnormal operation state alarm of electronic gastroscope,a new intelligent alarm method of abnormal operation state of electronic gastroscope is proposed.The normal or abnormal behavior of electronic gastroscope is described by the process of dual Markov embedding state,the normal state HHM model is established,the HMM learning is completed by Baum weltch algorithm,and the parameters of HMM model are estimated recursively,so that the re obtained HMM parameters gradually increase the posterior probability of the given observation sequence until convergence,and the effective interpretation of the training sample sequence is completed,to solve the existing evaluation problems,decoding problems and learning problems.In the process of abnormal intelligent alarm for the operation state of electronic gastroscope,the short sequence formed by sliding window is used to describe the current behavior of electronic gastroscope.According to the parameters of the hidden Markov model which describe the operation state of the normal electronic gastroscope,t he deviation of the sequence relative to the normal model behavior is measured by solving the output probability.The larger the deviation is,the greater the possibility of the behavior in the abnormal state is,and the abnormal intelligent alarm is realized.It is verified that the alarm of the proposed method is accurate and the overall performance is excellent.