基于外耳道气压的喉部健康状态特征分析

Extraction of pressure eigenvalues of external auditory canal based on wavelet packet decomposition

针对传统喉部检测存在侵入性、有创伤的缺点,提出基于外耳道气压的喉部健康状态检测方法。该方法基于外耳道气压与喉部状态的关联性,利用小波包分析方法对外耳道气压信号进行特征提取,结合机器学习和数据挖掘方法,能够智能地对喉部健康状态进行分类。以准确率94.4444%成功对17例输入实例进行分类,验证了外耳道气压与喉部健康状态的关联性。

Aiming at the invasive and traumatic shortcomings of traditional laryngeal detection methods,a method for detecting laryngeal health based on external auditory canal air pressure is proposed.Based on the correlation between the pressure of external auditory canal and the state of larynx,the method of wavelet packet analysis is used to extract the feature of pressure signal of external auditory canal,and the method of machine learning and data mining can intelligently classify the healthy state of larynx.Results 17 input cases were classified successfully with the accuracy rate 94.4444%,and the correlation between external auditory canal air pressure and laryngeal health was verified.